Differential diagnosis, methods and systems for mycosis fungoides

Abstract

This invention relates to a method for diagnosing mycosis fungoides (MF) or eczema and / or differentiating MF from eczema or psoriasis, the method comprising: determining the expression of at least one biomarker in a sample; differentiating MF from eczema and / or psoriasis based on the expression of the at least one biomarker in the sample; and generating a differential diagnosis result based on the expression of the at least one biomarker in the sample. The invention also relates to a system for diagnosing mycosis fungoides (MF) or eczema and / or differentiating MF from eczema or psoriasis, the system comprising: a processing component configured to output at least one dataset; and an analysis component configured to analyze the at least one dataset, wherein the analysis component comprises: a determination module configured to determine the expression of at least one biomarker in a sample; a differentiation module configured to differentiate MF from eczema and / or eczema or psoriasis based on the expression of the at least one biomarker in the sample; and a result generation module configured to generate a differential diagnosis result based on the expression of the at least one biomarker in the sample. Furthermore, the present invention relates to a kit for diagnosing eczema or mycosis fungoides, and / or a kit for distinguishing MF from eczema or psoriasis, said kit comprising at least one means for quantifying the expression of at least one biomarker in at least one sample.

Description

Technical Field

[0001] This invention relates to the field of medical diagnostics, specifically to the field of dermatological diagnostics. The object of this invention is to provide a diagnostic method for mycosis fungoides. More specifically, this invention provides a system for implementing the method and its corresponding uses.

[0002] introduction

[0003] Mycosis fungoides (MF) is a challenging and disabling disease characterized by malignant T-cell infiltration of the skin. As a subtype of cutaneous T-cell lymphoma (CTCL), MF is a rare malignant skin tumor with a broad spectrum of clinical presentations, ranging from erythematous and plaque stages to advanced tumor stages.

[0004] While current treatment options are feasible, they often have limited efficacy and are accompanied by significant side effects, highlighting the need for innovative treatment strategies. Furthermore, using conventional methods such as clinical observation and histopathology, these lymphomas are difficult to distinguish from eczema and / or psoriasis in their early stages.

[0005] Dobos et al. and Vermeer et al. published findings indicating that primary cutaneous lymphoma (CL) is a heterogeneous group of non-Hodgkin's lymphomas. Unlike its systemic counterpart, primary cutaneous T-cell lymphoma (CTCL) is more common than primary cutaneous B-cell lymphoma, accounting for over 80% of all primary cutaneous lymphomas. Although CTCL is a rare disease, it is of significant importance to public health programs, health insurance companies, and pharmaceutical manufacturers. The incidence of CTCL is approximately 1 in 100,000 per year, and it is estimated that CTCL is underdiagnosed, thus its incidence is underestimated. Patients with CTCL suffer from visible, itchy, and painful skin lesions, which affect their physical function, family life, social interactions, and interpersonal relationships. Diagnosis is often significantly delayed, which increases the psychosocial burden and leads to higher healthcare costs. Of patients diagnosed with early-stage CTCL, approximately one-quarter may progress to a more difficult-to-treat advanced form.

[0006] Assaf et al. published data showing that primary cutaneous lymphoma (CTCL) is a heterogeneous group of lymphomas that primarily affect the skin. The German Central Registry for Cutaneous Lymphoma (ZRKL) of the German Dermatology Society (DDG) assessed the epidemiology of cutaneous lymphoma in Germany, revealing that cutaneous T-cell lymphoma (85%) and cutaneous B-cell lymphoma (14%) are the most common. Mycosis fungoides is the most common manifestation of CTCL, accounting for 62% of all cases.

[0007] Walia et al. made groundbreaking molecular discoveries that advance CTCL research, revealing deeper insights into tumor biology and immune responses, and introducing targeted therapy strategies. Experts recommend using next-generation sequencing (NGS) for T-cell clonal detection, which helps identify major clonal sequences and manage minimal residual disease analysis. TCR γ-PCR (TCR gamma-PCR) is a standard method, but its sensitivity is inconsistent and its specificity is low.

[0008] A study by Lim et al. involving 246 patients with mycosis fungoides (MF) and Cezari syndrome (SS) found that 63% were male, with a median age at diagnosis of 49 years. Most patients were in the early stages, with 78.2% responding to treatment and 10.0% experiencing disease progression. The mean overall survival was 12.7 years, and the mortality rate was 2.5%. Prognostic factors associated with relapse-free survival included male sex, early disease stage, and failure to receive maintenance therapy after remission.

[0009] A study by Tsang et al. involving 1,981 patients with MF-CTCL found that severely ill patients incurred higher healthcare costs than those with mild to moderate cases. Approximately 51% of patients did not receive MF-CTCL-specific treatment within 60 days of diagnosis. Severe cases had a higher disease burden, healthcare costs, and healthcare resource utilization compared to mild to moderate cases. Low adherence to medication and high discontinuation rates may reflect post-treatment remission, but further research is needed.

[0010] Zhang et al.'s study, comparing early mycosis fungoides (eMF) lesions with healthy skin and benign inflammatory dermatitis, identified 349 differentially expressed genes in eMF lesions. Most of these genes showed upregulation in chronic dermatitis, making them less than ideal biomarkers for eMF. Two genes, TOX and PDCD1, demonstrated high discriminatory power between eMF lesions and benign dermatitis biopsy tissue. These genes, especially OX, could serve as molecular markers for the histological diagnosis of eMF, which currently presents a major diagnostic challenge.

[0011] McGirt et al. conducted a study comparing TOX staining in CTCL lineages, benign inflammatory skin diseases (such as psoriasis and cavernous dermatitis (BID)), and normal skin (NS). TOX-positive expression was detected in 73.6% of MF cases, compared to 31.6% in BID / NS. TOX expression also decreases during CTCL treatment, thus it may be a usable biomarker for MF.

[0012] Pileri et al. conducted a study investigating whether TOX could serve as a diagnostic or prognostic biomarker for patients with MF / SS. The results showed that TOX expression increased from the early to the late stages, but it was not a prognostic biomarker. This study concluded that TOX should be considered a prognostic biomarker rather than a diagnostic one.

[0013] Nielsen et al. presented a diagnostic classifier using TOX and TRAF1 that distinguished early melanoma (MF) from dermatitis with 85% accuracy in the discovery cohort and 80% accuracy in the independent validation cohort. TOX and TRAF1 protein levels were significantly elevated in early MF compared to the dermatitis group. TOX and TRAF1 levels also significantly increased during progression from early MF to tumor-stage MF. The protein expression levels of TOX and TRAF1 confirmed the difference between early MF and dermatitis, making it suitable for diagnosing MF.

[0014] Litvinov et al. (2017) addressed cutaneous T-cell lymphoma (CTCL), a rare but potentially devastating malignancy with an unclear pathogenesis. Early diagnosis takes up to six years, and the disease often masquerades as psoriasis or chronic eczema. A TruSeq study targeting RNA gene expression in 181 skin samples from CTCL patients and those affected by benign inflammatory skin conditions revealed significant molecular heterogeneity. Differential expression of genes such as TOX, FYB, LEF1, CCR4, ITK, EED, POU2AF, IL26, STAT5, BLK, GTSF1, and CCR4 may aid in the prediction of CTCL.

[0015] Litvinov et al. (2015) participated in a study aimed at predicting disease progression and stability in patients with mycosis fungoides and stage I-IV cutaneous T-cell lymphoma (CTCL). By analyzing gene expression in biopsy samples from 60 patients, the researchers identified three distinct clusters based on transcriptional profiling. The study found that 52 out of 240 genes could be categorized into expression patterns belonging to clusters 1-3, consistent with their suggested biological functions. Furthermore, 17 genes identified patients at risk of progression and differentiated mycosis fungoides / Cezari syndrome from benign mimicry diseases such as atopic dermatitis, unclassified dermatitis, and psoriasis. This study lays the foundation for developing personalized molecular approaches for the diagnosis and treatment of mycosis fungoides.

[0016] A study by Ralfkiaer et al. analyzed miRNA expression levels in 198 patients with CTCL, peripheral T-cell lymphoma, and benign skin diseases including psoriasis, atopic dermatitis, contact dermatitis, and unclassified dermatitis. The results showed that inducible and repressive miRNAs could distinguish CTCL from benign skin diseases with 90% accuracy. AqRT-PCR analysis confirmed differential expression of four miRNAs, demonstrating that miRNA classifiers with high specificity and sensitivity show great diagnostic potential in CTCL.

[0017] Soerensen et al. conducted a study to examine disease-specific miRNA expression in early mycosis fungoides erythematous and plaque stages compared to psoriasis. They found that the miRNA signatures exhibited in early mycosis fungoides overlapped with those in psoriasis. However, 39 miRNAs, including miR-142-3p, miR-150, and miR-146b, were specific to mycosis fungoides.

[0018] Nikolaou et al. conducted a retrospective analysis of all MF cases diagnosed and followed over 16 years, investigating the relationship between psoriasis and mycosis fungoides. They found that 7.8% (n=25 patients) met the inclusion criteria. Twenty patients had psoriatic lesions at the time of MF diagnosis, and in eight patients, typical histological features of both diseases were detected in the same biopsy sample, highlighting the necessity for accurate differential diagnosis.

[0019] As reported by Vaudreuil et al., mycosis fungoides not only presents as psoriasis or other diseases, but as they demonstrated in their case report, psoriasis can also present as mycosis fungoides.

[0020] Fahmy et al. reported a case of a patient diagnosed with mycosis fungoides (MF) after treatment with risankizumab (an IL-23 inhibitor) for psoriasis. Because of the established link between tumor necrosis factor-α inhibitors and cutaneous T-cell lymphoma (CTCL), and multiple reports describing the worsening of CTCL after exposure to other cytokine blockers, accurate diagnosis is crucial for differentiating new-onset MF in patients with inflammatory skin conditions from treatment-resistant lesions of benign skin diseases. Summary of the Invention

[0021] In view of the above, the object of the present invention is to overcome or at least mitigate the deficiencies and disadvantages of the prior art. More specifically, the object of the present invention is to provide a method for diagnosing eczema and mycosis fungoides, and a corresponding system for performing said method. Another object of the present invention is to provide a method for distinguishing eczema or psoriasis from mycosis fungoides, and a corresponding system for performing said method.

[0022] This invention achieves these objectives.

[0023] In a first aspect, the present invention relates to a method for diagnosing mycosis fungoides (MF) or eczema, the method comprising: determining the expression of at least one biomarker in a sample, distinguishing MF from eczema based on the expression of the at least one biomarker in the sample, and / or distinguishing MF from eczema or psoriasis, and generating a differential diagnosis result based on the expression of the at least one biomarker in the sample.

[0024] The method may include differentiating MF from eczema based on the expression of at least two biomarkers in the sample. The method may also include differentiating MF from eczema or psoriasis based on the expression of at least two biomarkers in the sample.

[0025] In one embodiment, the method may include distinguishing between MF and eczema based on the expression of at least three biomarkers in the sample, preferably based on at least four biomarkers in the sample, and more preferably based on at least five biomarkers in the sample.

[0026] In one embodiment, at least one of the at least one biomarkers may be selected from the group consisting of at least one of the following:

[0027] Furthermore, at least two of the at least one biomarker may be selected from the group consisting of at least one of the following:

[0028] These markers have unexpected advantages because they provide the method of the present invention with extremely high discriminative power. Surprisingly, HOMER1 has the highest discriminative power even when used alone, and its discriminative power is further improved by adding RNF213 and subsequently NLRC5.

[0029] In one embodiment, the at least one biomarker may optionally include at least one of the following:

[0030] In another embodiment, the at least one biomarker may include at least one of the following:

[0031] In other embodiments, the method can be used to diagnose early stages of MF or eczema, wherein the method may include identifying early stages of MF and eczema.

[0032] In addition, the samples may include samples collected from individuals.

[0033] The gene names used in the definition of the first aspect of this invention are all standard gene names recognized in the art. Table 1 below provides the full names and functional annotations of these genes.

[0034] Table 1. Genes that differentiate and / or isolate eczema and mycotic fibroids (MF)

[0035] Furthermore, the method may include: generating a skin condition status hypothesis based on the differential diagnosis result, predicting the individual's skin condition status based on the skin condition status hypothesis, and generating a skin condition status prediction. Additionally, this assessment step may be based on the differential diagnosis result. And or alternatively, this assessment step may also be based on the skin condition status prediction.

[0036] In one implementation, the method may include assessing whether the expression of at least one biomarker indicates that the individual may have MF or eczema.

[0037] In another embodiment, the method may include: determining an individual's skin condition status based on the skin condition threshold, wherein when the expression of at least one biomarker in the individual's sample is below the skin condition threshold, the method determines that the individual may have eczema; when the expression of at least one biomarker in the individual's sample is above the skin condition threshold, the method determines that the individual may have myxomatosis (MF).

[0038] Table 2. Evaluation of two different approaches: (1) using Lasso regression and (2) using forward feature selection in combination with different models, comparing the changes in the number of genes obtained and model performance.

[0039]

[0040] Furthermore, or additionally, the method may also include automatically generating at least one skin condition suggestion, wherein when the expression reading of the at least one biomarker in the individual sample is below the detection limit of the skin condition threshold, the at least one skin condition suggestion may include a prompt for evaluation by a medical professional.

[0041] In one implementation, the evaluation step may be based on the differential diagnosis result.

[0042] In another implementation, the assessment step may also be based on a prediction of the skin condition.

[0043] In addition, samples taken from individuals may include samples obtained from at least one individual suspected of having at least one of the following diseases and having skin lesions: eczema and cutaneous lymphoma.

[0044] In one embodiment, the method may include identifying at least one MF-eczema differential gene that is significantly differentially expressed in a sample among at least two differentially expressed genes.

[0045] In addition, the skin condition threshold may include at least one MF-eczema differentiation parameter and an MF-eczema differentiation reference parameter.

[0046] In other embodiments, the method may include measuring the fold change in the expression of at least one MF-eczema identifying gene. Furthermore, the measurement of the fold change may include measuring the log2 FoldChange order of change in the expression of at least one MF-eczema identifying gene. Additionally, the method may include processing at least one set of samples. And or additionally, processing the at least one set of samples may include: resampling the sample set at least 50 times, preferably at least 100 times, more preferably at least 150 times (such as 200 times); and automatically generating a resampled dataset.

[0047] In addition, processing the at least one set of samples may include performing sample cross-validation.

[0048] In one embodiment, the method may include performing differential gene expression analysis on at least one of the following: eczema versus non-lesion status, cutaneous lymphoma versus non-lesion status, and eczema versus cutaneous lymphoma.

[0049] In other embodiments, the method may include: identifying the at least one MF-eczema identifying gene, and identifying the at least one MF-eczema identifying gene as at least one of the at least one MF-eczema identifying genes associated with eczema, and at least one of the at least one MF-eczema identifying genes associated with cutaneous lymphoma, wherein the identification of the at least one MF-eczema identifying gene is based on a skin condition threshold.

[0050] For example, the method of the present invention may allow the selection of genes with an absolute value log2FoldChange > 1 and padj < 0.05, and the intersection of these genes that meet the absolute value condition with genes that distinguish between health and disease and distinguish between eczema and cutaneous lymphoma.

[0051] In one implementation, the method may include training at least one machine learning module using any data from any method step. Furthermore, the training may include using a training dataset that incorporates a normalization method and standard scaling. And or additionally, the normalization method may include performing M-value trimmed mean (TMM) normalization.

[0052] In addition, the method may include identifying at least one sample dataset for the prediction step.

[0053] In addition, in order to identify at least one sample dataset for the prediction step, the method may include: training a recursive forward feature selection (RFFS) using at least one MF-eczema identifying gene, and generating an optimized list of MF-eczema identifying genes based on the at least one sample dataset.

[0054] In one implementation, the method may include optimizing at least one of the following using RFFS: the skin condition threshold and the at least one MF-eczema differentiation parameter.

[0055] In another embodiment, the method may include storing a list of MF-eczema identifying genes in at least one of the following: random access memory (RAM), a server, a cloud, and a database.

[0056] In a further embodiment, the method may include: applying robust rank aggregation to the list of MF-eczema identifying genes to find at least one MF-eczema identifying gene that is consistently included in the list; applying Benjamin-Hochberg multiple test correction for the number of iterations and the number of MF-eczema identifying genes; and selecting at least one MF-eczema identifying gene from the list of MF-eczema identifying genes based on the results of robust rank aggregation and Benjamin-Hochberg correction.

[0057] In one embodiment, in order to generate a differential diagnosis result, the method may include: analyzing the expression of at least one biomarker in a sample and calculating the expression value.

[0058] In addition, the method may include analyzing the expression of at least one biomarker in a sample taken from at least one individual suffering from eczema.

[0059] In one embodiment, the method may include analyzing the expression of at least one biomarker in a sample taken from at least one individual suffering from mycosis fungoides.

[0060] Furthermore, the expression value can be calculated based on the analysis of the expression of at least one biomarker in samples taken from at least one individual with eczema, and the expression of at least one biomarker in samples taken from at least one individual with mycosis fungoides.

[0061] In other embodiments, the analysis and computation steps may be performed by the at least one machine learning module.

[0062] In a further embodiment, eczema may include at least one of the following: atopic eczema, contact dermatitis, pompholyx eczema, nummular eczema, allergic contact dermatitis, irritant dermatitis, seborrheic dermatitis, and stasis dermatitis.

[0063] In addition, the method may include: identifying at least one stage of the mycosis fungoides; and distinguishing the at least one stage, wherein the at least one stage may include at least one of the following: erythematous stage, plaque stage, tumor stage, and erythrodermic stage.

[0064] In one embodiment, the expression of at least one biomarker may include at least one expression level of both mRNA and protein. Furthermore, the expression of at least one mRNA level can be determined by at least one nucleic acid amplification method. And or additionally, the at least one nucleic acid amplification method may include at least one of the following: polymerase chain reaction (PCR) and isothermal amplification.

[0065] In addition, the PCR may include at least one of the following: conventional PCR, real-time PCR (qPCR), reverse transcription PCR (RT-PCR), nested PCR, multiplex PCR, digital PCR (dPCR), hot-start PCR, and droplet PCR.

[0066] In one embodiment, the isothermal amplification method may include at least one of the following: loop-mediated isothermal amplification (LAMP), recombinase polymerase amplification (RPA), helicase-dependent amplification (HDA), and nicking enzyme amplification reaction (NEAR).

[0067] In addition, loop-mediated isothermal amplification (LAMP) may include at least one of the following: reverse transcription LAMP (RT-LAMP), multiprimer LAMP (MPL), real-time LAMP, lateral flow chromatography LAMP, turbidity LAMP, and LAMP that combines clustered regularly spaced short palindromic repeats [CRISPR] (CRISPR-LAMP).

[0068] At least one expression at the protein level may include at least one antibody, said antibody including at least one of the following: a labeling antibody and a binding antibody.

[0069] In addition, the labeled antibody may include a label, which includes at least one of the following: a fluorescent label, a luminescent label, and a radioactive label.

[0070] In one embodiment, the method may include determining the at least one antibody using at least one labeled second antibody. Antibody detection may be performed using immunofluorescence assay and / or enzyme-linked immunosorbent assay (ELISA), or similar analytical techniques known in the art.

[0071] Furthermore, the method can be a computer-implemented method.

[0072] The present invention further relates to a method for distinguishing mycosis fungoides (MF) from eczema or psoriasis, the method comprising: determining the expression of at least one biomarker in a sample, distinguishing MF from eczema, and / or distinguishing MF from eczema or psoriasis.

[0073] In one embodiment, the method may include differentiating MF from eczema or psoriasis based on the expression of at least two and / or at least three biomarkers in the sample, preferably based on at least four biomarkers in the sample, and more preferably based on the expression of at least five biomarkers in the sample.

[0074] In one embodiment, at least one of the at least one biomarkers may be selected from the group consisting of at least one of the following:

[0075] In another embodiment, at least one of the at least one biomarker may be selected from the group consisting of at least one of the following:

[0076] Unless otherwise stated, the above description of methods for distinguishing between MF and eczema, using the cited biomarkers, also applies to methods for distinguishing between MF and eczema or psoriasis. In other words, the method can distinguish whether the data analyzed by the method is classified as MF or as eczema or psoriasis, where eczema or psoriasis refers to a single category containing at least one possibility of eczema or psoriasis. In the foregoing description of methods for distinguishing between MF and eczema, any mention of "eczema" is understood to also refer to "eczema or psoriasis".

[0077] The gene names used in the methods for distinguishing MF from eczema or psoriasis in this invention are all standard gene names recognized in the art. Table 3 below provides the full names and functional annotations of these genes.

[0078] Table 3. Differentiating and / or isolating genes from MF and eczema or psoriasis.

[0079]

[0080] In other embodiments, the psoriasis may consist of at least one of the following: plaque psoriasis, inverted psoriasis, guttate psoriasis, palmoplantar psoriasis, generalized pustular psoriasis, palmoplantar pustulosis, erythrodermic psoriasis, and scalp psoriasis.

[0081] The method of this invention is particularly advantageous because it allows for the handling of rare diseases with small sample sizes and imbalanced data, where the classification of a minority of categories is more important than that of the majority. Furthermore, the method of this invention has significant advantages in that it allows for the processing of high-dimensional feature spaces, the processing of highly correlated structures within high-dimensional feature spaces, the identification of clinically translatable concise features, and direct analysis at the gene space level, thereby achieving interpretability and clinical applicability, and filtering experimental noise such as low-expression genes.

[0082] In a second aspect, the present invention also relates to a system for diagnosing mycosis fungoides (MF) or eczema, the system comprising: a processing component configured to output at least one dataset; and an analysis component configured to analyze the at least one dataset, wherein the analysis component comprises: a determination module configured to determine the expression of at least one biomarker in a sample; a differentiation module configured to distinguish MF from eczema based on the expression of the at least one biomarker in the sample; and a result generation module configured to generate a differential diagnosis result based on the expression of the at least one biomarker in the sample.

[0083] In one implementation, the identification module may be configured to distinguish between MF and eczema based on the expression of at least two biomarkers in the sample.

[0084] In another embodiment, the identification module may be configured to distinguish between MF and eczema based on the expression of at least three biomarkers in the sample, preferably based on at least four biomarkers in the sample, and more preferably based on at least five biomarkers in the sample.

[0085] At least one of the biomarkers may be selected from the group consisting of at least one of the following:

[0086] At least two of the at least one biomarker may be selected from the group consisting of at least one of the following:

[0087] The at least one biomarker may optionally include at least one of the following:

[0088] In another embodiment, the at least one biomarker may include at least one of the following:

[0089] In one embodiment, the system may be configured to diagnose early MF or eczema, wherein the system may be configured to differentiate between early MF and eczema.

[0090] The samples may include samples collected from individuals.

[0091] In another embodiment, the system may be configured to: generate a skin condition status hypothesis based on the differential diagnosis result, predict the individual's skin condition status based on the skin condition status hypothesis, and generate a skin condition status prediction.

[0092] In addition, the system may include a prediction module configured to generate a hypothesis about the skin condition based on the differential diagnosis results.

[0093] In one implementation, the prediction module may be configured to predict the individual's skin condition based on the skin condition assumption and generate a skin condition prediction.

[0094] In another implementation, the prediction module may be integrated into the analysis component.

[0095] Furthermore, the system can be configured to assess skin condition based on the differential diagnosis results.

[0096] Furthermore, the system can be configured to predict and assess skin condition based on the skin condition status.

[0097] And or additionally, the system may be configured to assess whether the expression of at least one biomarker indicates that the individual may have MF or eczema.

[0098] In one implementation, the system may be configured to generate a skin condition threshold and determine the individual's skin condition status based on the skin condition threshold; When the expression of at least one biomarker in the individual sample is below the skin condition threshold, the system can be set to output that the individual may have eczema, and when the expression of at least one biomarker in the individual sample is above the skin condition threshold, the system can be set to output that the individual may have mycotic dermatitis (MF).

[0099] Furthermore, the system can be configured to automatically output at least one skin condition suggestion, wherein when the expression reading of at least one biomarker in the individual sample is below the detection limit of the skin condition threshold, the at least one skin condition suggestion includes a prompt for evaluation by a medical professional.

[0100] In one implementation, samples taken from an individual may include at least one individual suspected of having at least one of the following diseases and having skin lesions: eczema and cutaneous lymphoma.

[0101] Furthermore, the system can be configured to identify at least one MF-eczema differential gene that is significantly differentially expressed in a sample among at least two differentially expressed genes.

[0102] In one embodiment, the skin condition threshold may include at least one MF-eczema differentiation parameter and an MF-eczema differentiation reference parameter.

[0103] Furthermore, the system can be configured to measure the fold change in the expression of at least one MF-eczema-identifying gene.

[0104] In addition, the measurement of fold change may include measuring the log2 FoldChange magnitude change of at least one MF-eczema identifying gene.

[0105] In one embodiment, the system may be configured to process at least one set of samples.

[0106] In another embodiment, the system may be configured to resample the at least one set of samples at least 50 times, preferably at least 100 times, more preferably at least 150 times (such as 200 times), and automatically generate a resampled dataset.

[0107] In other embodiments, the system may be configured to process the at least one set of samples by performing sample cross-validation.

[0108] In another embodiment, the system may include performing differential gene expression analysis on at least one of the following: eczema versus non-lesion condition, cutaneous lymphoma versus non-lesion condition, and eczema versus cutaneous lymphoma.

[0109] Furthermore, the system can be configured to resample the at least one set of samples using an analysis component and automatically generate a resampled dataset.

[0110] Furthermore, the system may be configured to identify the at least one MF-eczema differential gene and to identify the at least one MF-eczema differential gene as at least one of the at least one MF-eczema differential genes associated with eczema, and at least one of the at least one MF-eczema differential genes associated with cutaneous lymphoma, wherein the system may be configured to determine the at least one MF-eczema differential gene based on a skin condition threshold.

[0111] In one implementation, the system may include at least one machine learning module.

[0112] In another embodiment, the system may include a training module configured to train the at least one machine learning module using data from any method steps.

[0113] Furthermore, the training module can be configured to train the at least one machine learning module using a training dataset that includes a normalization method and standard scaling. The normalization method may include performing M-value trimmed mean (TMM) normalization.

[0114] In other embodiments, the system may be configured to identify at least one sample dataset to predict the skin condition status.

[0115] In addition, the system may include and be configured to train a recursive forward feature selection (RFFS) using at least one MF-eczema identification gene and generate an optimized list of MF-eczema identification genes based on the at least one sample dataset.

[0116] In addition, the system may be configured to optimize at least one of the following using RFFS: the skin condition threshold and the at least one MF-eczema differentiation parameter.

[0117] Furthermore, or additionally, the system may be configured to store the list of MF-eczema identification genes in at least one of the following: random access memory (RAM), server, cloud, and database.

[0118] In one embodiment, the system may be configured to apply robust rank aggregation to a gene list to find at least one MF-eczema identifying gene that is consistently included in the list; apply Benjamin-Hochberg multiple test correction for the number of iterations and the number of MF-eczema identifying genes; and select at least one MF-eczema identifying gene from the list of MF-eczema identifying genes based on the results of robust rank aggregation and Benjamin-Hochberg correction.

[0119] Furthermore, or additionally, in order to generate differential diagnostic results, the system may be configured to analyze the expression of at least one biomarker in the sample and calculate the expression value.

[0120] Furthermore, the system can be configured to analyze the expression of at least one biomarker in samples taken from at least one individual suffering from eczema.

[0121] In one embodiment, the system may be configured to analyze the expression of at least one biomarker in a sample taken from at least one individual suffering from mycosis fungoides.

[0122] In another embodiment, the system may be configured to calculate expression values ​​based on the analysis of the expression of at least one biomarker in samples taken from at least one individual with eczema and the expression of at least one biomarker in samples taken from at least one individual with mycosis fungoides.

[0123] Furthermore, the system can be configured to analyze and calculate expression levels through the at least one machine learning module.

[0124] Eczema can include at least one of the following: atopic eczema, contact dermatitis, pompholyx eczema, nummular eczema, allergic contact dermatitis, irritant dermatitis, seborrheic dermatitis, and stasis dermatitis.

[0125] Furthermore, the system may be configured to identify at least one stage of mycosis fungoides and to differentiate the at least one stage, wherein the at least one stage may include at least one of the following: erythematous stage, plaque stage, tumor stage, and erythrodermic stage.

[0126] In one embodiment, the expression of at least one biomarker may include at least one of the following expression levels: mRNA expression and protein expression.

[0127] Furthermore, the expression level of at least one mRNA can be determined by at least one nucleic acid amplification method. The at least one nucleic acid amplification method may include at least one of the following: polymerase chain reaction (PCR) and isothermal amplification. The PCR may include at least one of the following: conventional PCR, real-time PCR (qPCR), reverse transcription PCR (RT-PCR), nested PCR, multiplex PCR, digital PCR (dPCR), hot-start PCR, and droplet PCR.

[0128] In addition, isothermal amplification methods may include at least one of the following: loop-mediated isothermal amplification (LAMP), recombinase polymerase amplification (RPA), helicase-dependent amplification (HDA), and nicking enzyme amplification reaction (NEAR). Loop-mediated isothermal amplification (LAMP) may include at least one of the following: reverse transcription LAMP (RT-LAMP), multiprimer LAMP (MPL), real-time LAMP, lateral flow chromatography LAMP, turbidity LAMP, and LAMP that binds clustered regularly spaced short palindromic repeats [CRISPR] (CRISPR-LAMP).

[0129] In one embodiment, at least one expression at the protein level may include at least one antibody, said antibody comprising at least one of the following: a labeled antibody and a binding antibody. The labeled antibody may include a marker, comprising at least one of the following: a fluorescent marker, a luminescent marker, and a radioactive marker.

[0130] In other embodiments, the system may be configured to determine the at least one antibody by using at least one labeled second antibody.

[0131] In one implementation, the system may be configured to perform any of the methods described herein.

[0132] In another embodiment, the system may include a computer-aided component configured to perform any of the methods described herein.

[0133] In addition, the method may include any steps that prompt the system described herein to perform the method described herein.

[0134] Furthermore, or additionally, the system may be an automated system.

[0135] The present invention also relates to a system for diagnosing mycosis fungoides (MF) with eczema or psoriasis, the system comprising: a processing component configured to output at least one dataset; and an analysis component configured to analyze the at least one dataset, wherein the analysis component comprises: a determination module configured to determine the expression of at least one biomarker in a sample; a differentiation module configured to differentiate MF from eczema based on the expression of the at least one biomarker in the sample; and a result generation module configured to generate a differential diagnosis result based on the expression of the at least one biomarker in the sample.

[0136] In one embodiment, the system may be configured to differentiate MF from eczema or psoriasis based on the expression of at least two and / or three biomarkers in the sample, preferably based on at least four biomarkers in the sample, and more preferably based on the expression of at least five biomarkers in the sample.

[0137] In one embodiment, at least one of the at least one biomarkers may be selected from the group consisting of at least one of the following:

[0138] In another embodiment, at least one of the at least one biomarker may be selected from the group consisting of at least one of the following:

[0139] Unless otherwise stated, the above description of the system for distinguishing between MF and eczema, using the cited biomarkers, is equally applicable to describing methods for distinguishing MF from eczema or psoriasis. In other words, the system can distinguish whether the data analyzed by the system is classified as MF or as eczema or psoriasis, where eczema or psoriasis refers to a single category containing at least one possibility of eczema or psoriasis. In the foregoing description of the system designed to distinguish between MF and eczema, any reference to "eczema" is understood to also refer to "eczema or psoriasis."

[0140] In other embodiments, the psoriasis may consist of at least one of the following: plaque psoriasis, inverted psoriasis, guttate psoriasis, palmoplantar psoriasis, generalized pustular psoriasis, palmoplantar pustulosis, erythrodermic psoriasis, and scalp psoriasis.

[0141] In a third aspect, the present invention relates to a kit for diagnosing eczema or mycosis fungoides, the kit comprising at least one means for quantifying the expression of at least one biomarker in at least one sample. The method may be performed according to the methods described herein. And or additionally, the method may be performed by the system described herein.

[0142] In one embodiment, the at least one means comprises at least one of the following: primers and antibodies. The at least one primer may include a forward primer and a reverse primer, and includes at least one of the following:

[0143] In addition, the kit may include at least one means for obtaining skin samples.

[0144] Furthermore, the kit may include at least one means for isolating RNA from skin samples. The at least one means for isolating RNA from skin samples may include at least one of the following: organic extraction, centrifugal column extraction, and magnetic bead extraction.

[0145] Furthermore, the kit may include at least one means for enriching RNA from skin samples. The at least one means for enriching RNA from skin samples may include at least one of the following: poly(A) enrichment, enzymatic removal, and probe-based depletion.

[0146] In one embodiment, the kit may include at least one skin analysis method. Analysis of the skin may include PCR.

[0147] In another embodiment, the kit may include at least one means for preparing tissue sections, including at least one of the following: a cryostat; a microstat; a non-invasive or minimally invasive sampling tool, such as a curette, tape, or patch; a biopsy puncture device; and a microneedle.

[0148] Furthermore, the kit may include a set of machine-readable instructions that, when executed, provide an authorized user with instructions to prompt the authorized user to perform at least one of the following: prompting the system according to any of the foregoing system embodiments to perform the method according to any of the foregoing method embodiments, prompting the user to perform the method according to any of the foregoing method embodiments, and using the kit according to any of the foregoing kit embodiments.

[0149] Furthermore, the at least one means for obtaining skin samples may include a microtome sectioning device for processing tape peels, skin scrapings, puncture biopsy tissue, micro or small biopsy tissue, and formaldehyde-fixed paraffin-embedded (FFPE) tissue.

[0150] Furthermore, the present invention relates to a kit for differentiating mycosis fungoides from eczema or psoriasis, the kit comprising at least one means for quantifying the expression of at least one biomarker in at least one sample. The method may be performed according to the methods described herein. And or additionally, the method may be performed by the system described herein.

[0151] In one embodiment, the at least one means comprises at least one of the following: primers and antibodies. The at least one primer may include a forward primer and a reverse primer, and includes at least one of the following:

[0152] Unless otherwise stated, the above description of a kit for distinguishing between MF and eczema, using the biomarkers employed in the corresponding method, also applies to kits for distinguishing between MF and eczema or psoriasis. In other words, the kit can distinguish whether data analyzed by the method is classified as MF or as eczema or psoriasis, where eczema or psoriasis refers to a single category containing at least one possibility of eczema or psoriasis. In the foregoing description of a kit designed to distinguish between MF and eczema, any mention of "eczema" is understood to also refer to "eczema or psoriasis".

[0153] In a fourth aspect, the present invention relates to the use of the system described herein in performing the methods described herein.

[0154] In addition, the use of the kit described herein in the methods described herein is also discussed.

[0155] Furthermore, in the use of the at least one biomarker in diagnosing eczema or MF based on samples taken from individuals, the at least one biomarker includes at least one of the following:

[0156] In one implementation, at least one primer comprising a forward primer and a reverse primer may include the following lengths:

[0157] Furthermore, the use of at least one biomarker in distinguishing MF from eczema or psoriasis in samples taken from an individual, said at least one biomarker comprising at least one of the following:

[0158] In one implementation, at least one primer comprising a forward primer and a reverse primer may include the following lengths:

[0159] The current technology is also described by the following numbered embodiments.

[0160] The following section discusses implementation methods. These implementation methods are abbreviated using the letter "M" followed by a number. All method implementation methods mentioned herein refer to these implementation methods.

[0161] M1. A method for diagnosing mycosis fungoides (MF) or eczema, the method comprising: Determine the expression of at least one biomarker in a sample. Differentiating between myasthenia gravis (MF) and eczema based on the expression of at least one biomarker in the samples, and Differential diagnostic results are generated based on the expression of at least one biomarker in the sample.

[0162] M2. The method according to the foregoing embodiments, wherein the method may include distinguishing MF from eczema based on the expression of at least two biomarkers in the sample.

[0163] M3. The method according to any one of the foregoing two embodiments, wherein the method may include distinguishing MF from eczema based on the expression of at least three biomarkers in the sample, preferably based on at least four biomarkers in the sample, more preferably based on at least five biomarkers in the sample.

[0164] M4. The method according to any one of the foregoing embodiments, wherein the at least one biomarker is selected from the group consisting of at least one of the following:

[0165] M5. The method according to any one of the foregoing embodiments, wherein at least two of the at least one biomarker are selected from the group consisting of at least one of the following:

[0166] M6. The method according to any one of the foregoing two embodiments, wherein the at least one biomarker optionally further comprises at least one of the following:

[0167] M7. The method according to any one of the foregoing three embodiments, wherein the at least one biomarker comprises at least one of the following:

[0168] M8. The method according to any one of the foregoing embodiments, wherein the method is used to diagnose MF or an early stage of eczema, wherein the method includes identifying MF and eczema in an early stage.

[0169] M9. The method according to the foregoing embodiments, wherein the sample includes a sample collected from an individual.

[0170] M10. The method according to any one of the foregoing embodiments, wherein the method includes

[0171] Based on the differential diagnosis results, a hypothesis about the skin condition is generated. Predict the individual's skin condition based on the assumed skin condition, and Generate skin condition predictions.

[0172] M11. The method according to the foregoing embodiments, wherein the evaluation step is based on the differential diagnosis results.

[0173] M12. The method according to any one of the foregoing two embodiments, wherein the evaluation step is based on the skin condition state prediction.

[0174] M13. The method according to any one of the foregoing embodiments, wherein the method includes assessing whether the expression of at least one biomarker indicates that the individual has MF or eczema.

[0175] M14. The method according to any one of the foregoing embodiments, wherein the method includes

[0176] Generate skin condition thresholds, and

[0177] The individual's skin condition is determined based on the aforementioned skin condition threshold. Wherein the expression of at least one biomarker in the individual sample is When the skin condition falls below the stated threshold, the method determines that the individual has eczema; and If the skin condition exceeds the stated threshold, the method determines that the individual has MF.

[0178] M15. The method according to any one of the foregoing two embodiments, wherein the method further comprises automatically generating at least one skin condition suggestion, wherein the at least one skin condition suggestion includes a suggestion for evaluation by a medical professional when the expression reading of the at least one biomarker in the individual sample is below the detection limit of the skin condition threshold.

[0179] M16. The method according to any one of the foregoing two embodiments, wherein the evaluation step is based on the differential diagnosis result.

[0180] M17. The method according to any one of the foregoing three embodiments and having the features of embodiment M10, wherein the evaluation step is based on the skin condition state prediction.

[0181] M18. The method according to the foregoing embodiments and having the features of embodiment M9, wherein the sample taken from an individual may include at least one individual suspected of having at least one of the following diseases and having skin lesions: eczema and cutaneous lymphoma.

[0182] M19. The method according to any one of the foregoing embodiments, wherein the method may include identifying at least one MF-eczema identification gene that is significantly differentially expressed in a sample among at least two differentially expressed genes.

[0183] M20. The method according to any of the foregoing embodiments and having the features of embodiment M14, wherein the skin condition threshold includes at least one MF-eczema differentiation parameter and an MF-eczema differentiation reference parameter.

[0184] M21. The method according to the two embodiments described above, wherein the method includes measuring the fold change in the expression of at least one MF-eczema identification gene.

[0185] M22. The method according to the foregoing embodiments, wherein the measurement of the fold change includes measuring the log2 FoldChange magnitude change of at least one MF-eczema identification gene.

[0186] M23. The method according to any one of the foregoing embodiments, wherein the method includes processing at least one set of samples.

[0187] M24. The method according to the foregoing embodiments, wherein processing the at least one set of samples includes

[0188] The sample set is resampled at least 50 times, preferably at least 100 times, more preferably at least 150 times, such as 200 times; and

[0189] Automatically generate resampled datasets.

[0190] M25. The method according to any one of the foregoing two embodiments, wherein processing the at least one set of samples includes performing sample cross-validation.

[0191] M26. A method according to any of the foregoing embodiments and having the features of any of embodiments M19 to M22, wherein the method includes performing differential gene expression analysis on at least one of the following: eczema and non-lesion conditions, cutaneous lymphoma and non-lesion conditions, and eczema and cutaneous lymphoma.

[0192] M27. A method according to any of the foregoing embodiments and having the features of embodiment M20, wherein the method includes...

[0193] Identify at least one of the MF-eczema differential genes, and

[0194] The at least one MF-eczema differential gene is used to distinguish it.

[0195] At least one of the at least MF-eczema differential genes associated with eczema, and

[0196] At least one of the at least MF-eczema differential genes is associated with cutaneous lymphoma. The differentiation of at least one MF-eczema identification gene is based on a skin condition threshold.

[0197] M28. The method according to any one of the foregoing embodiments, wherein the method includes training the at least one machine learning module using any data from any method step.

[0198] M29. The method according to the foregoing embodiments, wherein the training includes using a training dataset that includes a normalization method and a standard scaling.

[0199] M30. The method according to the foregoing embodiments, wherein the normalization method includes performing M-value trimmed mean (TMM) normalization.

[0200] M31. A method according to any one of the foregoing three embodiments and having the features of embodiment M10, wherein the method includes identifying at least one sample dataset for a prediction step.

[0201] M32. A method according to the foregoing embodiments and having the features of embodiment M19, wherein the method includes identifying at least one sample dataset for the prediction step.

[0202] Use at least one of the MF-eczema identification genes. Training Recursive Forward Feature Selection (RFFS), and An optimized list of genes for identifying MF-eczema is generated based on the at least one sample dataset.

[0203] M33. The method according to the foregoing embodiments and embodiments M14 and / or M20, wherein the method includes optimizing at least one of the following by means of RFFS: the skin condition threshold and the at least one MF-eczema differentiation parameter.

[0204] M34. The method according to any one of the foregoing two embodiments may include storing a list of MF-eczema identification genes in at least one of the following: random access memory (RAM), a server, a cloud, and a database.

[0205] M35. A method according to any of the foregoing embodiments and having the features of the foregoing embodiments, wherein the method includes...

[0206] Robust Rank Aggregation was applied to the list of MF-eczema diagnostic genes to find at least one MF-eczema diagnostic gene that is consistently included in the list. Benjamin-Hochberg multiple test corrections were applied to the number of iterations and the number of MF-eczema differential genes; and Based on the results of robust rank aggregation and Benjamin-Hochberg correction, at least one MF-eczema differential gene was selected from the list of MF-eczema differential genes.

[0207] M36. The method according to any one of the foregoing embodiments, wherein the method for generating differential diagnostic results includes...

[0208] Analyze the expression of at least one biomarker in the sample; and

[0209] Calculate the expression value.

[0210] M37. The method according to the foregoing embodiments, wherein the method may include analyzing the expression of at least one biomarker in a sample taken from at least one individual suffering from eczema.

[0211] M38. The method according to any one of the foregoing two embodiments, wherein the method may include analyzing the expression of at least one biomarker in a sample taken from at least one individual suffering from mycosis fungoides.

[0212] M39. The method according to the three embodiments described above, wherein the expression value is calculated based on the analysis of the expression of at least one biomarker in samples taken from at least one individual with eczema and the expression of at least one biomarker in samples taken from at least one individual with mycosis fungoides.

[0213] M40. A method according to any one of the foregoing four embodiments and having the features of embodiment M28, wherein the analysis and calculation steps are performed by the at least one machine learning module.

[0214] M41. The method according to any one of the foregoing embodiments, wherein the eczema includes at least one of the following: atopic eczema, contact dermatitis, pompholyx eczema, nummular eczema, allergic contact dermatitis, irritant dermatitis, seborrheic dermatitis, and stasis dermatitis.

[0215] M42. The method according to any one of the foregoing embodiments, wherein the method includes

[0216] Identify at least one stage of the mycosis fungoides; and

[0217] Distinguish between the at least one stage, The at least one stage includes at least one of the following: erythematous stage, plaque stage, tumor stage, and erythrodermic stage.

[0218] M43. The method according to any one of the foregoing embodiments, wherein the expression of the at least one biomarker includes at least one of the following expression levels: mRNA level expression, and Protein expression at the level.

[0219] M44. The method according to the foregoing embodiments, wherein at least one expression at the mRNA level is determined by at least one nucleic acid amplification method.

[0220] M45. The method according to the foregoing embodiments, wherein the at least one nucleic acid amplification method may include at least one of the following: polymerase chain reaction (PCR) and isothermal amplification.

[0221] M46. The method according to the foregoing embodiments, wherein the PCR includes at least one of the following: conventional PCR, real-time PCR (qPCR), reverse transcription PCR (RT-PCR), nested PCR, multiplex PCR, digital PCR (dPCR), hot-start PCR, and droplet PCR.

[0222] M47. The method according to any one of the foregoing two embodiments, wherein the isothermal amplification method includes at least one of the following: loop-mediated isothermal amplification (LAMP), recombinase polymerase amplification (RPA), helicase-dependent amplification (HDA), and nicking enzyme amplification reaction (NEAR).

[0223] M48. The method according to the foregoing embodiments, wherein the loop-mediated isothermal amplification (LAMP) includes at least one of the following: reverse transcription LAMP (RT-LAMP), multiprimer LAMP (MPL), real-time LAMP, lateral flow chromatography LAMP, turbidity LAMP, and LAMP that combines clustered regularly spaced short palindromic repeats [CRISPR] (CRISPR-LAMP).

[0224] M49. The method according to any one of the three preceding embodiments, wherein at least one expression at the protein level comprises at least one antibody, said antibody comprising at least one of the following: a labeling antibody and a binding antibody.

[0225] M50. The method according to the foregoing embodiments, wherein the labeled antibody includes a marker, which includes at least one of the following: a fluorescent marker, a luminescent marker, and a radioactive marker.

[0226] M51. The method according to any one of the foregoing two embodiments, wherein the method includes determining the at least one antibody by at least one labeled second antibody.

[0227] M52. The method according to any one of the foregoing embodiments, wherein the method is a computer-implemented method.

[0228] M53. The method according to any one of the foregoing embodiments, wherein the method is a method for distinguishing between MF and eczema or psoriasis, wherein in any of the foregoing method embodiments, the reference to "eczema" refers to "eczema or psoriasis".

[0229] M54. A method according to the foregoing method embodiments and having characteristics M1, M2 and / or M3, wherein the at least one biomarker is selected from the group consisting of at least one of the following:

[0230] M55. The method according to any one of the foregoing method embodiments, wherein at least two of the at least one biomarker are selected from the group consisting of at least one of the following:

[0231] M56. The method according to any one of the foregoing embodiments, wherein the psoriasis includes at least one of the following: plaque psoriasis, inverted psoriasis, guttate psoriasis, palmoplantar psoriasis, generalized pustular psoriasis, palmoplantar pustulosis, erythrodermic psoriasis, and scalp psoriasis.

[0232] The system implementations will be discussed below. These implementations are abbreviated with the letter "S" followed by a number. All system implementations mentioned in this document refer to these implementations.

[0233] S1. A system for diagnosing mycosis fungoides (MF) or eczema, said system comprising: Processing components, which are configured to output at least one dataset; and Analysis components, configured to analyze the at least one dataset, wherein the analysis components include: A determination module is configured to determine the expression of at least one biomarker in a sample; The identification module is configured to differentiate between MF and eczema based on the expression of at least one biomarker in the sample, and The results generation module is configured to generate differential diagnostic results based on the expression of at least one biomarker in the sample.

[0234] S2. The system according to the foregoing embodiments, wherein the identification module is configured to distinguish between MF and eczema based on the expression of at least two biomarkers in the sample.

[0235] S3. The system according to any one of the foregoing two embodiments, wherein the identification module is configured to distinguish MF from eczema based on the expression of at least three biomarkers in the sample, preferably based on at least four biomarkers in the sample, more preferably based on at least five biomarkers in the sample.

[0236] S4. The system according to any one of the foregoing system embodiments, wherein at least one of the at least one biomarker is selected from the group consisting of at least one of the following:

[0237] S5. The system according to any one of the foregoing system embodiments, wherein at least two of the at least one biomarker are selected from the group consisting of at least one of the following:

[0238] S6. The system according to any one of the foregoing two embodiments, wherein the at least one biomarker optionally further comprises at least one of the following:

[0239] S7. The system according to any one of the foregoing three embodiments, wherein the at least one biomarker comprises at least one of the following:

[0240] S8. The system according to any one of the foregoing system embodiments, wherein the system is configured to diagnose MF or eczema in an early stage, wherein the system is configured to identify MF and eczema in an early stage.

[0241] S9. The system according to the foregoing embodiments, wherein the sample includes a sample collected from an individual.

[0242] S10. The system according to any one of the foregoing embodiments, wherein the system is configured to

[0243] Based on the differential diagnosis results, a hypothesis about the skin condition is generated. Based on the aforementioned skin condition assumptions, predict the individual's skin condition status, and Generate skin condition predictions.

[0244] S11. A system according to any of the foregoing system embodiments and having the features of the foregoing embodiments, wherein the system includes a prediction module configured to generate a skin condition state hypothesis based on the differential diagnosis result.

[0245] S12. A system according to any one of the foregoing two embodiments and having the features of the foregoing embodiments, wherein the prediction module is configured as follows:

[0246] Based on the aforementioned skin condition assumptions, predict the individual's skin condition status, and

[0247] Generate skin condition predictions.

[0248] S13. The system according to any one of the foregoing two embodiments, wherein the prediction module is integrated into the analysis component.

[0249] S14. The system according to the foregoing embodiments, wherein the system is configured to assess skin condition based on the differential diagnosis results.

[0250] S15. The system according to any one of the foregoing two embodiments, wherein the system is configured to assess skin condition based on the skin condition state prediction.

[0251] S16. The system according to any one of the foregoing system embodiments, wherein the system is configured to assess whether the expression of at least one biomarker indicates that the individual has MF or eczema.

[0252] S17. The system according to any one of the foregoing system embodiments, wherein the system is configured to

[0253] Generate skin condition thresholds, and

[0254] The individual's skin condition is determined based on the aforementioned skin condition threshold. Wherein the expression of at least one biomarker in the individual sample is When the skin condition falls below the stated threshold, the system is configured to output that the individual has eczema; and When the skin condition exceeds the threshold, the system is set to output that the individual has MF.

[0255] S18. The system according to any one of the foregoing two embodiments, wherein the system is configured to automatically output at least one skin condition suggestion, wherein the at least one skin condition suggestion includes a prompt for evaluation by a medical professional when the expression reading of the at least one biomarker in the individual sample is below the detection limit of the skin condition threshold.

[0256] S19. The system according to the foregoing embodiments and having the features of embodiment S9, wherein the sample taken from an individual includes at least one individual suspected of having at least one of the following diseases and having skin lesions: eczema and cutaneous lymphoma.

[0257] S20. The system according to any one of the foregoing embodiments, wherein the system is configured to identify at least one MF-eczema identification gene that is significantly differentially expressed in a sample among at least two differentially expressed genes.

[0258] S21. The system according to any of the foregoing embodiments and having the features of embodiment S17, wherein the skin condition threshold includes at least one MF-eczema differentiation parameter and an MF-eczema differentiation reference parameter.

[0259] S22. The system according to the two embodiments described above, wherein the system is configured to measure the fold change in the expression of at least one MF-eczema identification gene.

[0260] S23. The system according to the foregoing embodiments, wherein the measurement of the fold change includes measuring the log2 FoldChange magnitude change of at least one MF-eczema identification gene.

[0261] S24. The system according to any one of the foregoing system embodiments, wherein the system is configured to process at least one set of samples.

[0262] S25. The system according to any one of the foregoing system embodiments, wherein the system is configured to

[0263] At least 50 resampling operations are performed on at least one set of samples, preferably at least 100, more preferably at least 150, such as 200; and

[0264] Automatically generate resampled datasets.

[0265] S26. The system according to any one of the foregoing two embodiments, wherein the system is configured to process the at least one set of samples by performing sample cross-validation.

[0266] S27. The system according to any of the foregoing embodiments and having any of the features of embodiments S20 to S22, wherein the system includes performing differential gene expression analysis on at least one of the following: eczema and non-lesion status, cutaneous lymphoma and non-lesion status, and eczema and cutaneous lymphoma.

[0267] S28. The system according to any one of the foregoing two embodiments, wherein the system is configured to resample the at least one set of samples by means of an analysis component and automatically generate a resampled dataset.

[0268] S29. The system according to any of the foregoing system embodiments and having the features of embodiment S21, wherein the system is configured as follows:

[0269] Identify at least one of the MF-eczema differential genes, and

[0270] The at least one MF-eczema differential gene is used to distinguish it.

[0271] At least one of the at least MF-eczema differential genes associated with eczema, and

[0272] At least one of the at least MF-eczema differential genes is associated with cutaneous lymphoma. The system is configured to distinguish at least one MF-eczema identification gene based on the skin condition threshold.

[0273] S30. The system according to any one of the foregoing system implementation methods, wherein the system includes at least one machine learning module.

[0274] S31. The system according to any one of the foregoing system embodiments, wherein the system includes a training module configured to train the at least one machine learning module using any data from any method step.

[0275] S32. The system according to any one of the foregoing system embodiments, wherein the training module is configured to train the at least one machine learning module using a training dataset that includes a normalization method and a standard scaling.

[0276] S33. The system according to the foregoing embodiments, wherein the normalization method includes performing M-value trimmed mean (TMM) normalization.

[0277] S34. The system according to any one of the foregoing four embodiments and having the features of embodiment S10, wherein the system is configured to identify at least one sample dataset to predict the skin condition state.

[0278] S35. The system according to the foregoing embodiments and having the features of embodiment S29, wherein the system includes and is configured to

[0279] Use at least one of the MF-eczema identification genes. Training Recursive Forward Feature Selection (RFFS), and An optimized list of genes for identifying MF-eczema is generated based on the at least one sample dataset.

[0280] S36. The system according to the foregoing embodiments and embodiment S21, wherein the system is configured to optimize at least one of the following by means of RFFS: the skin condition threshold and the at least one MF-eczema differentiation parameter.

[0281] S37. The system according to any one of the foregoing two embodiments, wherein the system is configured to store the list of MF-eczema identification genes in at least one of the following: random access memory (RAM), server, cloud, and database.

[0282] S38. The system according to any of the foregoing system embodiments and having the features of the foregoing embodiments, wherein the system is configured as follows:

[0283] Robust Rank Aggregation was applied to the gene list to find at least one MF-eczema differential gene that was consistently included in the list; Benjamin-Hochberg multiple test corrections were applied to the number of iterations and the number of MF-eczema differential genes; and Based on the results of robust rank aggregation and Benjamin-Hochberg correction, at least one MF-eczema differential gene was selected from the list of MF-eczema differential genes.

[0284] S39. The system according to any one of the foregoing system embodiments, wherein the system for generating differential diagnostic results is configured to...

[0285] Analyze the expression of at least one biomarker in the sample; and

[0286] Calculate the expression value.

[0287] S40. The system according to the foregoing embodiments, wherein the system is configured to analyze the expression of at least one biomarker in a sample taken from at least one individual suffering from eczema.

[0288] S41. The system according to any one of the foregoing two embodiments, wherein the system is configured to analyze the expression of at least one biomarker in a sample taken from at least one individual suffering from mycosis fungoides.

[0289] S42. The system according to the three embodiments described above, wherein the system is configured to calculate expression values ​​based on analyzing the expression of at least one biomarker in samples taken from at least one individual suffering from eczema and the expression of at least one biomarker in samples taken from at least one individual suffering from mycosis fungoides.

[0290] S43. The system according to any one of the foregoing four embodiments and having the features of embodiment S30, wherein the system is configured to analyze and calculate expression levels through the at least one machine learning module.

[0291] S44. The system according to any one of the foregoing system embodiments, wherein the eczema includes at least one of the following: atopic eczema, contact dermatitis, pompholyx eczema, nummular eczema, allergic contact dermatitis, irritant dermatitis, seborrheic dermatitis, and stasis dermatitis.

[0292] S45. The system according to any one of the foregoing system embodiments, wherein the system is configured to

[0293] Identify at least one stage of the mycosis fungoides; and

[0294] Distinguish between the at least one stage. The at least one stage includes at least one of the following: erythematous stage, plaque stage, tumor stage, and erythrodermic stage.

[0295] S46. The system according to any one of the foregoing system embodiments, wherein the expression of the at least one biomarker includes at least one of the following expression levels: mRNA level expression, and Protein expression at the level.

[0296] S47. The system according to the foregoing embodiments, wherein at least one expression at the mRNA level is determined by at least one nucleic acid amplification method.

[0297] S48. The system according to the foregoing embodiments, wherein the at least one nucleic acid amplification method includes at least one of the following: polymerase chain reaction (PCR) and isothermal amplification system.

[0298] S49. The system according to the foregoing embodiments, wherein the PCR includes at least one of the following: conventional PCR, real-time PCR (qPCR), reverse transcription PCR (RT-PCR), nested PCR, multiplex PCR, digital PCR (dPCR), hot-start PCR, and droplet PCR.

[0299] S50. The system according to any one of the foregoing two embodiments, wherein the isothermal amplification method comprises at least one of the following: loop-mediated isothermal amplification system (LAMP), recombinase polymerase amplification (RPA), helicase-dependent amplification system (HDA), and nicking enzyme amplification reaction (NEAR).

[0300] S51. The system according to the foregoing embodiments, wherein the loop-mediated isothermal amplification (LAMP) includes at least one of the following: reverse transcription LAMP (RT-LAMP), multiprimer LAMP (MPL), real-time LAMP, lateral flow chromatography LAMP, turbidity LAMP, and LAMP that combines clustered regularly spaced short palindromic repeats [CRISPR] (CRISPR-LAMP).

[0301] S52. The system according to any one of the three embodiments described above, wherein at least one expression at the protein level includes at least one antibody, said antibody including at least one of the following: a labeling antibody and a binding antibody.

[0302] S53. The system according to the foregoing embodiments, wherein the labeled antibody includes a marker, which includes at least one of the following: a fluorescent marker, a luminescent marker, and a radioactive marker.

[0303] S54. The system according to any one of the foregoing two embodiments, wherein the system is configured to determine the at least one antibody by at least one labeled second antibody.

[0304] S55. The system according to any one of the foregoing system embodiments, wherein the system is configured to perform the method steps according to any one of the foregoing method embodiments.

[0305] S56. The system according to any one of the foregoing system embodiments, wherein the system includes a computer-aided component configured to perform any one of the method steps according to any one of the foregoing embodiments.

[0306] M53. The method according to any one of the foregoing method embodiments, wherein the method includes prompting the system according to any one of the foregoing system embodiments to perform any step of the method according to any one of the foregoing method embodiments.

[0307] S57. The system according to any one of the foregoing system embodiments, wherein the system is an automated system.

[0308] S58. The system according to any one of the foregoing system embodiments, wherein the system is a system for distinguishing between MF and eczema or psoriasis, wherein in any of the foregoing system embodiments, any reference to "eczema" refers to "eczema or psoriasis".

[0309] S59. The system according to the foregoing system implementation and having features S1, S2 and / or S3, wherein at least one of the at least one biomarkers is selected from the group consisting of at least one of the following:

[0310] S60. The system according to any one of the foregoing system embodiments, wherein at least two of the at least one biomarker are selected from the group consisting of at least one of the following:

[0311] S61. The system according to any one of the foregoing system embodiments, wherein the psoriasis includes at least one of the following: plaque psoriasis, inverted psoriasis, guttate psoriasis, palmoplantar psoriasis, generalized pustular psoriasis, palmoplantar pustulosis, erythrodermic psoriasis, and scalp psoriasis.

[0312] The following section discusses kit implementation methods. These implementation methods are abbreviated with the letter "K" followed by a number. All kit implementation methods mentioned in this document refer to these implementation methods.

[0313] K1. A kit for diagnosing eczema or mycosis fungoides, the kit comprising at least one means for quantifying the expression of at least one biomarker in at least one sample.

[0314] K2. The kit according to the foregoing embodiments, wherein the method is according to any one of the foregoing method embodiments.

[0315] K3. The kit according to any one of the foregoing two embodiments, wherein the method is performed via a system according to any one of the foregoing system embodiments.

[0316] K4. The kit according to any one of the foregoing kit embodiments, wherein at least one means comprises at least one of the following: primers and antibodies.

[0317] K5. The kit according to the foregoing embodiments, wherein the at least one primer comprises a forward primer and a reverse primer, and includes at least one of the following:

[0318] K6. The kit according to any one of the foregoing kit embodiments, wherein the kit includes at least one means for obtaining skin samples.

[0319] K7. The kit according to any one of the foregoing kit embodiments, wherein the kit includes at least one means for isolating RNA from skin samples.

[0320] K8. The kit according to any one of the foregoing embodiments, wherein the at least one means of isolating RNA from skin samples includes at least one of the following: organic extraction, centrifugal column extraction, and magnetic bead extraction.

[0321] K9. The kit according to any one of the foregoing kit embodiments, wherein the kit includes at least one means for enriching RNA from skin samples.

[0322] K10. The kit according to any one of the foregoing embodiments, wherein the at least one means of enriching RNA from skin samples includes at least one of the following: poly(A) enrichment, enzymatic removal, and probe-based depletion.

[0323] K11. The kit according to any one of the foregoing kit embodiments, wherein the kit includes at least one skin analysis method.

[0324] K12. The kit according to the foregoing embodiments, wherein the analysis of the skin includes PCR.

[0325] K13. The kit according to any one of the foregoing kit embodiments, wherein the kit comprises at least one means for preparing tissue sections: a cryostat; a microstat; a non-invasive or minimally invasive sampling tool, such as a curette, tape, patch; a biopsy puncture device; and a microneedle.

[0326] K14. The kit according to any one of the foregoing kit embodiments, wherein the kit includes a set of machine-readable instructions that, when executed, provide an authorized user with instructions to prompt the authorized user to perform at least one of the following:

[0327] The system described in any of the foregoing system implementations shall execute the method described in any of the foregoing method implementations. The prompt indicates that the method described according to any of the foregoing method implementations shall be executed, and Use the kit described according to any of the foregoing kit implementation methods.

[0328] K15. The kit according to any of the foregoing kit embodiments and having the features of embodiment K5, wherein the at least one means for obtaining skin samples includes a microtome sectioning device for processing tape peels, skin scrapings, puncture biopsy tissue, micro or small biopsy tissue, and formalin-fixed paraffin-embedded (FFPE) tissue.

[0329] K16. The kit according to any one of the foregoing kit embodiments, wherein the kit is a kit for a method of distinguishing eczema or psoriasis from mycosis fungoides, wherein in any of the foregoing kit embodiments, the term "eczema" refers to "eczema or psoriasis".

[0330] K17. The kit according to the aforementioned kit embodiment and having the K4 feature, wherein the at least one primer comprises a forward primer and a reverse primer, and comprises at least one of the following:

[0331] The following section discusses the implementation methods. These implementation methods are abbreviated with the letter "U" followed by a number. All implementation methods mentioned herein refer to these implementation methods.

[0332] U1. The system according to any one of the foregoing system embodiments is used to perform the method according to any one of the foregoing method embodiments.

[0333] U2. Use of the kit according to any one of the foregoing kit embodiments in the method according to any one of the foregoing method embodiments.

[0334] U3. Use of the at least one biomarker in diagnosing eczema or MF based on samples taken from an individual, wherein the at least one biomarker includes at least one of the following:

[0335] U4. The use of the at least one biomarker in differentiating MF from eczema or psoriasis based on samples taken from individuals, the at least one biomarker comprising at least one of the following:

[0336] The present invention will now be described with reference to the accompanying drawings, which illustrate embodiments of the invention. These embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0337] Figure 1 A system for diagnosing MF or eczema and / or differentiating MF from eczema or psoriasis according to an embodiment of the present invention is illustrated schematically. Figure 2 A schematic depiction of differential gene expression analysis for feature selection according to an embodiment of the present invention is shown. Figure 3 The illustration depicts the use of Lasso regression instead of forward selection according to an embodiment of the present invention; Figure 4A graphical representation of a disease-significant gene box plot for diagnosing MF or eczema according to an embodiment of the present invention is shown. Figure 5 The concept describes the correlation between Lasso logistic regression and FFS biomarkers according to embodiments of the present invention; Figure 6 A gene for the method according to an embodiment of the present invention is schematically depicted; Figure 7 The detection of biomarkers according to embodiments of the present invention is illustrated schematically; Figure 8 Box plots depicting the measured expression of genes HOMER1, RNF213, GBP4, and LCK according to embodiments of the present invention are shown. Figure 9 Box plots depicting the measured expression of genes HOMER1, RNF213, NLRC5, GBP4, ZC3H12D, and LCK according to embodiments of the present invention are shown. Figure 10 Box plots depicting the measured expression of genes HOMER1, RNF213, NLRC5, GBP4, ZC3H12D, LCK, FLRT5, and PNLIPRP3 according to embodiments of the present invention are shown. Figure 11 A computing device suitable for performing the method according to an embodiment of the present invention is schematically depicted.

[0338] It should be noted that not all accompanying drawings have all reference numerals. Instead, in some drawings, certain reference numerals are omitted for the sake of brevity and simplicity. Embodiments of the invention will now be described with reference to the accompanying drawings.

[0339] Figure 1 A system 100 for diagnosing mycosis fungoides (MF) or eczema (and / or eczema or psoriasis) according to an embodiment of the present invention is schematically depicted. In simple terms, system 100 is used to diagnose mycosis fungoides (MF) or eczema (and / or eczema or psoriasis) and includes a processing component 110 configured to output at least one dataset, and an analysis component 120 configured to analyze said at least one dataset.

[0340] The analysis component 120 may further include a plurality of modules, such as a determination module 122 configured to determine the expression of at least one biomarker in a sample; a differentiation module 124 configured to differentiate MF from eczema (and / or eczema or psoriasis) based on the expression of at least one biomarker in a sample; and a result generation module 126 configured to generate a differential diagnosis result based on the expression of at least one biomarker in a sample.

[0341] In one embodiment, system 100 may further include a server (not depicted), which may be a local server or a remote server, or a server that is partly local and partly remote. In other embodiments, the server may also be configured in the cloud.

[0342] In addition, the identification module 124 can be configured to identify MF from eczema (and / or eczema or psoriasis) based on the expression of at least two biomarkers in the sample.

[0343] Furthermore, system 100 may include a prediction module (not depicted) configured to generate a skin condition status hypothesis based on the differential diagnosis results. In one embodiment, the prediction module may be configured to predict the individual's skin condition status based on the skin condition status hypothesis and generate a skin condition status prediction. In another embodiment, the prediction module may be integrated into analysis component 120.

[0344] Figure 2 The differential gene expression analysis according to an embodiment of the present invention is illustrated schematically. In short, Figure 2 The training dataset used by the method of the present invention is described, wherein the method may include processing at least one set of samples. In one embodiment of the invention, processing the at least one set of samples may include performing multiple steps, specifically multiple steps of a computer-implemented method. For example, this may include resampling the set of samples at least 50 times, preferably at least 100 times, more preferably at least 150 times, such as 200 times; and automatically generating a resampled dataset. Furthermore, processing the at least one set of samples may include performing sample cross-validation. Additionally, as... Figure 2 The method described may include performing differential gene expression analysis on at least one of the following: eczema versus non-lesion conditions, cutaneous lymphoma versus non-lesion conditions, and eczema versus cutaneous lymphoma. To this end, the method may include: identifying the at least one MF-eczema differential gene and distinguishing the at least one MF-eczema differential gene into at least one MF-eczema differential gene associated with eczema, and at least one MF-eczema differential gene associated with cutaneous lymphoma, wherein the distinction of the at least one MF-eczema differential gene is based on a skin condition threshold. Figure 2The training is further described, which includes using a training dataset containing a normalization method and standard scaling, wherein the normalization method may include performing M-value trimmed mean (TMM) normalization. To identify at least one sample dataset for the prediction step, the method may include: training a recursive forward feature selection (RFFS) using at least one MF-eczema identifying gene, and generating an optimized list of MF-eczema identifying genes based on the at least one sample dataset. This method allows for feature selection, which may subsequently allow for forward sequential feature selection based on logistic regression, and / or sorting of features according to their addition time. Furthermore, as... Figure 2 The method described may include: applying robust rank aggregation to the list of MF-eczema identifying genes to find at least one MF-eczema identifying gene that is consistently included in the list; applying Benjamin-Hochberg multiple test correction for the number of iterations and the number of MF-eczema identifying genes; and selecting at least one MF-eczema identifying gene from the list of MF-eczema identifying genes based on the results of robust rank aggregation and Benjamin-Hochberg correction.

[0345] in other words, Figure 2 Several steps are schematically disclosed, including applying differential gene expression analysis between eczema / cutaneous lymphoma patients and all non-lesion patients; filtering the two results for genes with an absolute log2FoldChange value >= 1 and padj < 0.01 in the comparison of eczema or cutaneous lymphoma (or both), these genes are the distinguishing genes between healthy and unhealthy; initiating bootstrapping, which can be repeated, for example, 200 times; drawing samples from 6 patients, such as 3 eczema patients and 3 cutaneous lymphoma patients; performing differential gene expression analysis between eczema and cutaneous lymphoma; extracting genes with an absolute log2FoldChange value > 1 and padj < 0.05; taking the intersection of genes that meet the absolute value condition with genes from the filtering step, i.e., genes that distinguish between healthy and disease and genes that distinguish between eczema and cutaneous lymphoma; using other samples from the training set and applying TMM normalization and standard scaling; using only the genes from the intersection and training recursive forward feature selection, optimizing the F1 score and storing the genes in a list. Furthermore, Figure 2 The method describes applying robust rank aggregation to a list of 200 genes to find genes that are consistently contained in the list; applying Benjamin-Hochberg correction for multiple testing based on the number of iterations and the number of genes; and selecting genes with padj values ​​below 0.01.

[0346] also, Figure 2The forward sequential feature selection is described. This can be performed by using the F1 score as the performance metric; using a tolerance of 0.5%; using a logistic regression model (with weights to balance the imbalanced dataset and L2 regularization); using hierarchical cross-validation with 4 folds (3 folds for training, 1 fold for evaluation); starting with the maximum performance (in this case, the F1 score) where the sum of the empty feature set is 0; adding a gene to the feature set and training the model on the training fold; evaluating the performance (F1 score) on the reserved fold of the cross-validation; storing the average performance across all folds; iterating through all genes and comparing performance; if the performance is better than the maximum performance minus the tolerance, adding the gene that achieves the best performance to the gene set; setting the maximum performance to that performance; and repeating all the described steps until the maximum performance no longer increases.

[0347] like Figure 2 The method shown can also be applied to forward feature selection using several different models (such as SVM, logistic regression, xgboost); however, the results of this method are consistent across different methods.

[0348] Figure 3 The illustration schematically depicts the use of Lasso regression instead of forward selection according to an embodiment of the present invention. In simple terms, Figure 3 Depicting and Figure 2 The same concept is shown, but Lasso logistic regression is used instead of forward selection. Generally, there are two approaches to finding biomarkers: (1) finding the minimum biomarker that yields good predictive performance, and (2) outputting more genes and reflecting a more complete disease profile, which can achieve better predictive performance due to the larger number of genes. Therefore, Figure 2 and Figure 3 The difference lies in the use of Lasso logistic regression.

[0349] Simply put, Figure 3The method is illustrated schematically and includes the following steps: applying differential gene expression (DGE) analysis between eczema / cutaneous lymphoma patients and all non-lesion patients; filtering the two results for genes with an absolute value of log2FoldChange >= 1 and padj < 0.01 in the comparison of eczema or cutaneous lymphoma (or both), these genes are subsequently distinguishing between healthy and unhealthy genes; initiating a bootstrap sampling method repeated multiple times (e.g., 200 times); drawing samples from 6 patients (3 eczema patients and 3 cutaneous lymphoma patients); performing DGE analysis between eczema and cutaneous lymphoma; extracting genes with an absolute value of log2FoldChange > 1 and padj < 0.05; taking the intersection of genes with absolute values ​​with genes from the filtering step, i.e., genes distinguishing between healthy and disease and genes distinguishing between eczema and cutaneous lymphoma; using other samples from the training set and applying TMM normalization and standard scaling; training a logistic regression model with L1 regularization and λ=1; and ranking the genes according to the absolute value of the gene coefficients.

[0350] Discard genes with a coefficient of 0; store the genes in a list; apply robust rank aggregation to the list of 200 genes to find genes that are consistently contained in the list; apply Benjamin-Hochberg correction for multiple tests based on the number of iterations and the number of genes; select genes with a padj value below 0.01; remove genes with a variance inflation factor (VIF) > 5. This can be achieved using the following formula:

[0351] It allows the calculation of the VIF of feature j.

[0352] Example 1: Evaluation

[0353] This invention includes evaluating data to diagnose cutaneous T-cell lymphoma (MF) or eczema. For this purpose, a label "0" can be used to encode cutaneous T-cell lymphoma, and a label "1" can be used to encode eczema. Therefore, genes can be evaluated using nested hierarchical cross-validation, where hyperparameter tuning can be performed in the inner loop and the fitted model evaluated in the outer loop. For this, an f2 score can be used and pos_label = 0 can be set to minimize the number of cutaneous lymphoma patients classified as eczema patients. Furthermore, this invention allows concatenating the prediction results and ground truth data for each sample to obtain the prediction results and ground truth data for the entire training set, and calculating the evaluation metric across the entire training set. This yields more meaningful results because each sample contains only a small number of MF samples. To obtain a robust estimate of the metric, the above process can be repeated multiple times (e.g., 100 times) with different splits (different random seeds), and the mean and standard deviation can be reported. For example, a logistic regression model or a linear support vector machine (SVM) can be used.

[0354] The following are the metrics used for evaluation in this invention:

[0355] The above metrics can be used to evaluate a logistic regression model trained using nested cross-validation on three previously selected genes, following the above method.

[0356]

[0357] In addition, external test sets can be used for evaluation, the reliability of which may depend on the size of the test set. Here are some examples of such external evaluations:

[0358] Example 2: Discovered biomarkers and their evaluation using Lasso regression

[0359] The table above shows, for example Figure 3 The evaluation results of the method shown. Figure 4 Box plots depicting significant genes used to diagnose MF or eczema in different diseases were presented. Specifically, Figure 4 A bilateral Wilcoxon rank-sum test was performed, revealing significant differences in the expression of the identified biomarkers between eczema and cutaneous T-cell lymphoma. (Tagged "...") "This corresponds to an adjusted p-value less than 0.0001."

[0360] Example 3: Performing FFS using logistic regression

[0361] Tables 4.1 and 4.2 below show the results when all 36 genes are protected, such as Figure 2 The results of the method are shown in Table 4.1. Table 4.2 shows that it is possible to detect the minimum feature combination with three genes, and it performs well. Table 4.2 shows that feature combinations with four genes can also be found, but the performance is much worse than in Table 4.1. In particular, the ability to classify MF (F1 score) is significantly reduced.

[0362] Table 4.1: HOMER1, NLRC5, RNF213

[0363] Table 4.2: SLFN12L, BTN3A1, SLMAF8, ZC3H12D

[0364] Example 4: Applying Lasso logistic regression and VIF

[0365] Tables 4.1 and 4.2 below show, for example... Figure 3 The result of the method. Specifically, the method of the present invention will be described herein according to... Figure 3 As shown, this was applied to a dataset where all 36 genes were protected.

[0366] The method of this invention can find smaller combinations of features, namely the nine biomarkers in Table 4.2 compared to the 23 biomarkers in Table 4.1. However, its performance appears to be worse than when using the 23 identified genes. Notably, the performance using these nine genes is even worse than using... Figure 2 The method used to identify the three genes (HOMER1, NLRC5, and RNF213) was even worse.

[0367] Table 5.1: CALCRL, GRAP2, TNFRSF10B, PRXL2A, H1-1, FLRT3, USP20, ENAH, VOPP1, GBP4, DISC1, LRATD1, S100A11, CDK1, CSRP2, ALOX15B, MUC16, SH2D1A, MMP1, RCSD1, MT-ND3, CES1, PNLIPRP3

[0368] Table 5.2: TRAF1, KRT6B, FCGR2B, MFNG, FADS2, ZAP70, SLOC2B1, PRC1, BTN3A1

[0369] Figure 5 A conceptual description is provided of the correlation between Lasso logistic regression and FFS biomarkers according to embodiments of the present invention. In short, Figure 5 The high correlation between NLRC5 and RNF213 was depicted. Specifically, Figure 5 The method of the present invention is described (specifically, respectively) Figure 2 and Figure 3 The method described above detects pairwise correlations between genes. Figure 5 The results showed a significant high correlation between RNF213 and NLRC5.

[0370] Example 5: Applying FFS to Logistic Regression

[0371] Tables 6.1 and 6.2 below show, for example... Figure 2 The results of the method are shown in Table 6.1. Specifically, Table 6.1 shows the performance of the model comprising three genes in this invention, and how it performs when applied. Figure 3 What happens when the method is used to protect these three genes? The method still finds a small feature set containing only four genes. However, in terms of performance, these genes cannot compare to those previously detected. In particular, the F1 score for MF drops significantly. Table 6.2 specifically shows GBP4, which is included in all 36 genes, as well as SLFN12L and MLLT6, which were also selected when all 36 genes were removed.

[0372] Table 6.1: HOMER1, NLRC5, RNF213

[0373] Table 6.2: BTN3A1, GBP4, SLFN12L MLLT6

[0374] Figure 6 A biomarker for the method according to an embodiment of the present invention is schematically depicted. In short, Figure 6The frameworks from which all 36 protected genes originated are depicted. FFS_Logistic, FFS_SVM, and FFS_Xgboost are results produced by performing the steps of the method of this invention, in which different models, such as logistic regression, SVM, and Xgboost, are used in the multivariate module. Lasso includes a method for applying variance inflation factor before... Figure 3 The method of this invention shown contains all 29 genes detected, while lasso_vif includes the results after applying the variance inflation factor. Notably, genes GBP4, CXCL9, IKZF3, and NLRC5 were detected in both frameworks of this invention.

[0375] Example 6: Applying Lasso logistic regression and VIF

[0376] Tables 7.1 and 7.2 below show, for example... Figure 3 The results of the method can be observed, specifically, when applied... Figure 3 When the method is described (which always extracts a larger combination of disease features), it can be found that it is not affected by protecting only the three genes HOMER1, NLRC5, and RNF213, because these genes were not selected previously.

[0377] Table 7.1: CALCRL, GRAP2, TNFRSF10B, PRXL2A, H1-1, FLRT3, USP20, ENAH, VOPP1, GBP4, DISC1, LRATD1, S100A11, CDK1, CSRP2, ALOX15B, MUC16, SH2D1A, MMP1, RCSD1, MT-ND3, CES1, PNLIPRP3

[0378] Table 7.2: CALCRL, GRAP2, TNFRSF10B, PRXL2A, H1-1, FLRT3, USP20, ENAH, VOPP1, GBP4, DISC1, LRATD1, S100A11, CDK1, CSRP2, ALOX15B, MUC16, SH2D1A, MMP1, RCSD1, MT-ND3, CES1, PNLIPRP3

[0379] Figure 7 The detection of biomarkers according to embodiments of the present invention is illustrated schematically. Specifically, Figure 7The invention describes the detection of biomarkers according to the method of the present invention, wherein the detected genes depend on which model is used in the multivariate module of the present invention, such as logistic regression, SVM, and Xgboost, as detailed in the appendix. Figure 2 As stated above. Furthermore... Figure 7 It shows that HOMER1 will be consistently selected regardless of which model is chosen. Furthermore, Figure 7 The results show that RNF213 and NLRC5 are also selected by SVM and logistic regression models.

[0380] Example 7: Permutations and combinations of HOMER1, RNF213, and NLRC5

[0381] Tables 8.1, 8.2, and 8.3 show the individual permutation results for HOMER1, RNF213, and NLRC5, respectively. The individual performance of each gene can be observed, significantly indicating that HOMER1 is the best-performing gene when used alone.

[0382] Table 8.1: HOMER1

[0383] Table 8.2: RNF213

[0384] Table 8.3: NLRC5

[0385] Tables 8.4, 8.5, and 8.6 show the results of the two-gene permutations of HOMER1, RNF213, and NLRC5. Specifically, it can be seen that HOMER1 and RNF213 performed best in all these permutations and may be able to distinguish between MF and eczema individually (in the absence of NLRC5).

[0386] Table 8.4: HOMER1 and RNF213

[0387] Table 8.5: HOMER1 and NLRC5

[0388] Table 8.6: RNF213 and NLRC5

[0389] Furthermore, the method of this invention can also achieve the ranking of biomarker performance. As mentioned earlier, HOMER1 is the most important gene in terms of performance. This is also reflected in... Figure 2In the robust rank aggregation output of the method, HOMER1 has the smallest p-value. Using the p-values ​​of the robust rank aggregation, the order after HOMER1 will be NLRC5, followed by RNF213.

[0390] However, as mentioned above, when performing dual-gene permutations, the performance improvement brought by the combination of HOMER1 and RNF213 is greater than that of the combination of HOMER1 and NLRC5.

[0391] Specifically, based on the results in Table 7.1, the genes can be sorted according to the p-value as follows: (1) HOMER1, (2) NLRC5, and (3) RNF213. Based on the results in Table 7.2, the genes can be sorted according to the p-value as follows: (1) HOMER1, (2) RNF213, and (3) NLRC5, where NLRC5 is optional.

[0392] Table 8.7: Using HOMER1, NLRC5, and RNF213

[0393] Table 8.8: Using HOMER1 and RNF213

[0394] This invention also includes evaluation data to distinguish MF from eczema or psoriasis. For this purpose, a label "0" can be used to encode cutaneous T-cell lymphoma, and a label "1" can be used to encode eczema or psoriasis. Therefore, genes can be evaluated using nested hierarchical cross-validation, where hyperparameters can be applied to fine-tune the inner loop and the fitted model is evaluated in the outer loop. For this, the f2 score can be used and pos_label=0 can be set to minimize the number of cutaneous lymphoma patients classified as eczema or psoriasis patients. Furthermore, this invention allows concatenating the prediction results and ground truth data for each sample to obtain the prediction results and ground truth data for the entire training set, and calculating evaluation metrics across the entire training set. This yields more meaningful results because each sample contains only a small number of MF samples.

[0395] The following are the metrics used for evaluation in this invention:

[0396] The above metrics can be used to evaluate multiple models trained on the selected genes as described above using nested cross-validation.

[0397] It is worth noting that, Figure 8 , Figure 9 and Figure 10 The results of the evaluation of qPCR data demonstrate the real-time application of the invention. This evaluation was performed based on two reference genes: TBP and SDHAF.

[0398] Figure 8 Box plots depict the performance of measured genes HOMER1, RNF213, GBP4, and LCK analyzed using logistic regression models and linear support vector machines. Model performance using these genes was measured based on model type, selected reference gene, F1 score, sensitivity, specificity, ROC AUC, and equilibrium accuracy.

[0399] Figure 9 Box plots depict the performance of measured genes HOMER1, RNF213, NLRC5, GBP4, ZC3H12D, and LCK analyzed using logistic regression models and linear support vector machines. Model performance using these genes was measured based on model type, selected reference gene, F1 score, sensitivity, specificity, ROC AUC, and equilibrium accuracy.

[0400] Figure 10 Box plots depict the performance of measured genes HOMER1, RNF213, NLRC5, GBP4, ZC3H12D, LCK, FLRT5, and PNLIPRP3 analyzed using logistic regression models and linear support vector machines. Model performance using these genes was measured based on model type, selected reference gene, F1 score, sensitivity, specificity, ROC AUC, and equilibrium accuracy.

[0401] Figure 11 A schematic diagram of a computing device 200 is provided. The computing device 200 may include a computing unit 35, a first data storage unit 30A, a second data storage unit 30B, and a third data storage unit 30C.

[0402] The computing device 200 can be a single computing device or a combination of computing devices. The computing device 200 can be deployed locally or remotely (such as in a cloud solution).

[0403] Different data can be stored on different data storage units 30. Additional data storage can also be provided, and / or the aforementioned storage units can be combined at least partially.

[0404] The computing unit 35 can access the first data storage unit 30A, the second data storage unit 30B, and the third data storage unit 30C through the internal communication channel 260 (which may include a bus connection 260).

[0405] The computing unit 30 may be a single processor or multiple processors, and may be, but is not limited to, a CPU (Central Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), APU (Accelerated Processing Unit), ASIC (Application-Specific Integrated Circuit), ASIP (Application-Specific Instruction Set Processor), or FPGA (Field-Programmable Gate Array). The first data storage unit 30A may be a single or multiple, and may be, but is not limited to, volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or phase-change RAM (P-RAM).

[0406] The second data storage unit 30B can be single or multiple, and can be, but is not limited to, volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or phase-change RAM (P-RAM).

[0407] The third data storage unit 30C can be single or multiple, and can be, but is not limited to, volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or phase-change RAM (P-RAM).

[0408] It should be understood that the first data storage unit 30A (also known as the encryption key storage unit 30A), the second data storage unit 30B (also known as the data share storage unit 30B), and the third data storage unit 30C (also known as the decryption key storage unit 30C) can typically be part of the same memory. That is, each device can provide only one general-purpose data storage unit 30, which can be configured to store the corresponding encryption key (such that the area of ​​the data storage unit 30 storing the encryption key is the encryption key storage unit 30A), the corresponding data element share (such that the area of ​​the data storage unit 30 storing the data element share is the data share storage unit 30B), and the corresponding decryption key (such that the area of ​​the data storage unit 30 storing the decryption key is the decryption key storage unit 30A).

[0409] In some embodiments, the third data storage unit 30C may be a secure storage device 30C, such as a self-encrypting memory or a hardware-based full-disk encryption memory, which can automatically encrypt all stored data. Data can only be decrypted from the storage component after a party requesting access to the third data storage unit 30C has successfully authenticated; this party may be a user, a computing device, a processing unit, etc. In some embodiments, the third data storage unit 30C can only be connected to the computing unit 35, and the computing unit 35 may be configured to never output data received from the third data storage unit 30C. This ensures the secure storage and processing of the encryption key (i.e., private key) stored in the third data storage unit 30C.

[0410] In some embodiments, the second data storage unit 30B may not be provided; instead, the computing device 200 may be configured to receive corresponding encrypted shares from the database 60. In some embodiments, the computing device 200 may include the second data storage unit 30B and may be configured to receive corresponding encrypted shares from the database 60.

[0411] The computing device 200 may include additional storage components 240, which may be single or multiple, and may be, but are not limited to, volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or parametric RAM (P-RAM). Storage component 240 may also be connected to other components of the computing device 200, such as computing component 35, via internal communication channel 260.

[0412] Furthermore, the computing device 200 may include an external communication component 230. The external communication component 230 may be configured to facilitate the transmission and / or reception of data to and / or from external devices (e.g., backup devices, recovery devices, databases). The external communication component 230 may include antennas (e.g., Wi-Fi antennas, NFC antennas, 2G / 3G / 4G / 5G antennas, etc.), USB ports / plugs, LAN ports / plugs, contacts providing electrical connections, etc. The external communication component 230 may transmit and / or receive data based on a communication protocol, which may include instructions for transmitting and / or receiving data. These instructions may be stored in the storage component 240 and executed by the computing unit 35 and / or the external communication component 230. The external communication component 230 may be connected to the internal communication component 260. Therefore, data received by the external communication component 230 may be provided to the storage component 240, the computing unit 35, the first data storage unit 30A and / or the second data storage unit 30B and / or the third data storage unit 30C. Similarly, the data stored in storage component 240, first data storage unit 30A and / or second data storage unit 30B and / or third data storage unit 30C, as well as the data generated by computing unit 35, can be provided to external communication component 230 for transmission to external devices.

[0413] In addition, the computing device 200 may include an input user interface 210, which allows a user of the computing device 200 to provide at least one input (e.g., an instruction) to the computing device 200. For example, the input user interface 210 may include buttons, a keyboard, a touchpad, a mouse, a touch screen, a joystick, etc.

[0414] In addition, the computing device 200 may also include an output user interface 220, which allows the computing device 200 to provide instructions to the user. For example, the output user interface 210 may be an LED, a display, a speaker, etc.

[0415] The input and output user interface 200 can also be connected to the internal components of the device 200 via the internal communication component 260.

[0416] The processor can be single or multiple, and can be, but is not limited to, a CPU, GPU, DSP, APU, or FPGA. The memory can be single or multiple, and can be, but is not limited to, volatile or non-volatile, such as SDRAM, DRAM, SRAM, flash memory, MRAM, F-RAM, or P-RAM.

[0417] The data processing apparatus may include data processing means, such as processor units, hardware accelerators, and / or microcontrollers. Data processing apparatus 20 may include storage components, such as main memory (e.g., RAM), cache (e.g., SRAM), and / or auxiliary memory (e.g., HDD, SDD). The data processing apparatus may include a bus configured to facilitate data exchange between components of the data processing apparatus (e.g., communication between storage components and processing components). The data processing apparatus may include a network interface card configured to connect the data processing apparatus to a network, such as the Internet. The data processing apparatus may include a user interface, such as: (1) Output user interface, such as: - A screen or display set up to show visual data (e.g., a graphical user interface that displays a questionnaire to a user). - A speaker configured to transmit audio data (e.g., play audio data to a user). (2) Input user interface, such as: - A camera configured to capture visual data (e.g., capturing images and / or videos of the user). - A microphone configured to capture audio data (e.g., record the user's audio). - Keyboards configured to allow the insertion of text and / or other keyboard commands (e.g., allowing users to input text data and / or other keyboard commands by enabling users to type on the keyboard) and / or touchpads, mice, touchscreens, joysticks - are configured to facilitate navigation between different graphical user interfaces of the questionnaire.

[0418] The data processing device may be a processing unit configured to execute program instructions. The data processing device may be a system-on-a-chip (SoC) including a processing unit, storage components, and a bus. The data processing device may be a personal computer, laptop computer, PDA, smartphone, or tablet computer. The data processing device may be a server, whether local or remote. The data processing device may be a processing unit or a SoC that can be connected to a personal computer, laptop computer, PDA, smartphone, tablet computer, and / or a user interface (such as the aforementioned user interface).

[0419] It should be noted that not all accompanying figures have all the figure reference numerals. On the contrary, in some figures, some figure reference numerals are omitted for the sake of simplicity and clarity.

[0420] The reference numerals and letters appearing in parentheses within the claims to identify features described in the embodiments and illustrated in conjunction with the accompanying drawings are provided merely as examples to help the reader understand the claimed matters. The inclusion of these reference numerals and letters should not be construed as setting any limitation on the scope of the claims.

[0421] Although preferred embodiments have been described above with reference to the accompanying drawings, those skilled in the art will understand that these embodiments are for illustrative purposes only and should not be construed as limiting the scope of the invention, which is defined by the claims.

[0422] Whenever relative terms such as “about,” “substantially,” or “approximately” are used in this specification, the term should also be interpreted to include the exact term. That is, for example, “substantially straight” should be interpreted to also include “(exactly) straight.”

[0423] Whenever steps are described in the foregoing or appended claims, it should be noted that the order of steps described in this document may be nonspecific. That is, unless otherwise specified or clearly understood by a person skilled in the art, the order of steps may be accidental. In other words, when this document states, for example, that a method includes steps (A) and (B), this does not necessarily mean that step (A) precedes step (B), that step (A) may be performed (at least partially) simultaneously with step (B), or that step (B) is performed before step (A). Furthermore, when it is mentioned that step (X) precedes another step (Z), this does not mean that there are no steps between steps (X) and (Z). That is, step (X) preceding step (Z) covers the case where step (X) is performed directly before step (Z), and also covers the case where one or more steps (Y1) are performed after step (X) is performed… followed by step (Z). The same consideration applies when terms such as “after” or “before” are used.

[0424] Scientific literature cited in this article

[0425] (1) Dobos, G., Pohrt, A., Ram-Wolff, C., Lebbé, C., Bouaziz, JD, Battistella, M., ... and Masson, AD (2020). Epidemiology of cutaneous T-celllymphomas: a systematic review and meta-analysis of 16,953 patients. Cancers, 12(10), 2921.

[0426] (2) Vermeer, M. (2021). Epidemiology of cutaneous lymphoma. The British journal of dermatology, 184(6), 993.

[0427] (3) Dobos, G., Miladi, M., Michel, L., Ram-Wolff, C., Battistella, M., Bagot, M., and de Masson, A. (2022). Recent advances on cutaneous lymphoma epidemiology. La Presse Médicale, 51(1), 104 - 108.

[0428] (4) Assaf, C., Gellrich, S., Steinhoff, M., Nashan, D., Weiße, F., Dippel, E.,... and Sterry, W. (2007). Cutaneous lymphomas in Germany: an analysis of the Central Cutaneous Lymphoma Registry of the German Society of Dermatology (DDG). JDDG: Journal der Deutschen Dermatologischen Gesellschaft, 5(8), 662 - 668

[0429] (5) Walia, R., and Yeung, C. (2020). An update on molecular biology of cutaneous T cell lymphoma. Frontiers in oncology, 9, 505501.

[0430] (6) Lim, H. L. J., Tan, E. S. T., Tee, S. I., Ho, Z. Y., Boey, J. J.J., Tan, W. P.,... and Tan, S. H. (2019). Epidemiology and prognostic factors for mycosis fungoides and Sézary syndrome in a multi‐ethnic Asian cohort: a 12‐year review. Journal of the European Academy of Dermatology and Venereology, 33(8), 1513-1521.

[0431] (7) Tsang, Y., Gu, T., Sharma, G., Raspa, S., Drake III, B., and Tan, H. (2018). Healthcare resource utilization, costs of care, and treatment of mycosis fungoides cutaneous T-cell lymphoma patterns in a large managed care population: a retrospective US claims-based analysis. Journal of Dermatological Treatment , 29 (8), 747-753.

[0432] (8) Zhang, Y., Wang, Y., Yu, R., Huang, Y., Su, M., Xiao, C.,... and Zhou, Y. (2012). Molecular markers of early-stage mycosis fungoides. Journal of investigative dermatology , 132 (6), 1698-1706.

[0433] (9) McGirt, L. Y., Degesys, C. A., Johnson, V. E., Zic, J. A., Zwerner, J. P., and Eischen, C. M. (2016). TOX expression and role in CTCL. Journal of the European Academy of Dermatology and Venereology , 30 (9), 1497 - 1502.

[0434] (10) Pileri, A., Cavicchi, M., Bertuzzi, C., Righi, S., Zengarini, C., Sabattini, E.,... and Agostinelli, C. (2022). TOX expression in mycosis fungoides and sezary syndrome. Diagnostics , 12 (7), 1582.

[0435] (11) Nielsen, P. R., Eriksen, J. O., Lindahl, L. M., Wehkamp, U., Andersen, G. R., Bzorek, M.,... and Gjerdrum, L. M. R. (2019). Diagnostic 2 - Gene Classifier in Early - Stage Mycosis Fungoides: A Retrospective Multicenter Study. Blood , 134 , 2772.

[0436] (12) Litvinov, I. V., Tetzlaff, M. T., Thibault, P., Gangar, P., Moreau, L., Watters, A. K.,... and Duvic, M. (2017). Gene expression analysis in Cutaneous T - Cell Lymphomas (CTCL) highlights disease heterogeneity and potential diagnostic and prognostic indicators. Oncoimmunology , 6(5), e1306618.

[0437] (13) Litvinov, I. V., Netchiporouk, E., Cordeiro, B., Doré, M. A., Moreau, L., Pehr, K.,... and Kupper, T. S. (2015). The use of transcriptional profiling to improve personalized diagnosis and management of cutaneous T-cell lymphoma (CTCL). Clinical Cancer Research, 21(12), 2820-2829.

[0438] (14) Ralfkiaer, U., Hagedorn, P. H., Bangsgaard, N., Løvendorf, M. B., Ahler, C. B., Svensson, L.,... and Odum, N. (2011). Diagnostic microRNA profiling in cutaneous T-cell lymphoma (CTCL). Blood, The Journal of the American Society of Hematology , 118 (22), 5891-5900.

[0439] (15) S.T. Sørensen, T. Litman, M. Gluud, P. Celis, S. Torres-Rusillo, A. Willerslev-Olsen, etc., MIRNA signature in early-stage mycosis fungoides, Acta Dermato-Venereologica. 102 (2022). doi:10.2340 / actadv.v102.628.

[0440] (16) V. Nikolaou, L. Marinos, E. Moustou, E. Papadavid, A. Economidi, E. Christofidou, etc., Psoriasis in patients with mycosis fungoides: A clinicopathological study of 25 patients, Journal of the European Academy of Dermatology and Venereology. 31 (2017) 1848–1852. doi:10.1111 / jdv.14365.

[0441] (17) A.M. Vaudreuil, C.M. Stroud, S. Hsu, Psoriasis mimicking mycosis fungoides clinically, Dermatology Online Journal. 23 (2017). doi:10.5070 / d3235034934.

[0442] (18) L.M. Fahmy, C.M. Schreidah, B.A. Lapolla, C.M. Magro, L.J. Geskin, Mycosis fungoides diagnosed after exposure to risankizumab for psoriasis, JAAD Case Reports. 41 (2023) 85–89. doi:10.1016 / j.jdcr.2023.08.049.

Claims

1. A method for diagnosing mycosis fungoides (MF) or eczema, the method comprising: Determine the expression of at least one biomarker in a sample. To differentiate between myasthenia gravis (MF) and eczema based on the expression of at least one biomarker in the samples, and Differential diagnostic results are generated based on the expression of at least one biomarker in the sample. The at least one of the biomarkers is selected from the group consisting of at least one of the following:

2. The method according to the preceding claim, wherein the at least one biomarker comprises at least one of the following:

3. A method for distinguishing mycosis fungoides (MF) from eczema or psoriasis, wherein in any of the preceding method claims and / or any method claim having the features of claim 1, reference to "eczema" refers to "eczema or psoriasis," wherein at least one of the at least one biomarkers is selected from the group consisting of at least one of the following:

4. The method according to any one of the preceding claims, wherein the method comprises Based on the differential diagnosis results, a hypothesis about the skin condition is generated. Predict the individual's skin condition based on the assumed skin condition, and Generate skin condition predictions.

5. The method according to any one of the preceding claims, wherein the method comprises Generate skin condition thresholds, and The individual's skin condition is determined based on the aforementioned skin condition threshold. Wherein the expression of at least one biomarker in the individual sample When the skin condition falls below the stated skin condition threshold, the method determines that the individual has eczema. as well as If the skin condition exceeds the stated threshold, the method determines that the individual has Mycotic Flavouring (MF). The method further includes automatically generating at least one skin condition suggestion, wherein when the expression reading of the at least one biomarker in the individual sample is below the detection limit of the skin condition threshold, the at least one skin condition suggestion includes a prompt for evaluation by a medical professional.

6. The method according to any one of the preceding claims, wherein the method comprises identifying at least one MF-eczema differential gene that is significantly differentially expressed in a sample from at least two differentially expressed genes, wherein the skin condition threshold comprises at least one MF-eczema differentiation parameter and an MF-eczema differentiation reference parameter, wherein the method comprises measuring the fold change in expression of at least one MF-eczema differential gene, wherein the measurement of the fold change comprises measuring the log2 FoldChange order of change of at least one MF-eczema differential gene. The method includes Identify at least one of the MF-eczema differential genes, and The at least one MF-eczema differential gene is used to distinguish it. At least one of the at least MF-eczema differential genes associated with eczema, and At least one of the at least MF-eczema differential genes is associated with cutaneous lymphoma. The differentiation of at least one MF-eczema identification gene is based on a skin condition threshold.

7. The method according to any one of the preceding claims, wherein the method includes training at least one machine learning module using any data from any step of the method, wherein the training includes using a training dataset that includes a normalization method and a standard scaling, wherein the normalization method includes performing M-value trimmed mean (TMM) normalization, and wherein the method includes identifying at least one sample dataset for the prediction step.

8. The method according to any one of the preceding claims, wherein the method comprises Identify at least one stage of the mycosis fungoides; and Distinguish between the at least one stage, The at least one stage includes at least one of the following: erythematous stage, plaque stage, tumor stage, and erythrodermic stage.

9. A system for diagnosing mycosis fungoides (MF) or eczema, said system comprising: A processing component that is configured to output at least one dataset; as well as Analysis components, configured to analyze the at least one dataset, wherein the analysis components include: A determination module is configured to determine the expression of at least one biomarker in a sample; The identification module is configured to differentiate between MF and eczema based on the expression of at least one biomarker in the sample, and The results generation module is configured to generate differential diagnostic results based on the expression of at least one biomarker in the sample. At least one of the biomarkers mentioned above is selected from the group consisting of:

10. The system according to the preceding claim, wherein the at least one biomarker comprises at least one of the following:

11. A system for distinguishing mycosis fungoides (MF) from eczema or psoriasis, wherein in any of the preceding system claims and / or any system claim having the features of claim 9, reference to "eczema" refers to "eczema or psoriasis," wherein the at least one biomarker is selected from the group consisting of at least one of the following:

12. The system according to any one of the preceding two claims, wherein the system is configured to Based on the differential diagnosis results, a hypothesis about the skin condition is generated. Based on the aforementioned skin condition assumptions, predict the individual's skin condition status, and Generate skin condition predictions.

13. The system according to any one of claims 9 to 12, wherein the system is configured to Generate skin condition thresholds, and The individual's skin condition is determined based on the aforementioned skin condition threshold. Wherein the expression of at least one biomarker in the individual sample is When the skin condition falls below the stated threshold, the system is configured to output that the individual has eczema; and When the skin condition exceeds the stated threshold, the system is set to output that the individual has MF; The system is configured to automatically output at least one skin condition suggestion, wherein the at least one skin condition suggestion includes a prompt for evaluation by a medical professional when the expression reading of at least one biomarker in the individual sample is below the detection limit of the skin condition threshold.

14. The system according to any one of claims 9 to 13, wherein the system is configured to identify at least one differentially expressed MF-eczema differential gene in the sample from at least two differently expressed genes, wherein the skin condition threshold includes at least one MF-eczema differential parameter and an MF-eczema differential reference parameter, wherein the system is configured to measure the fold change in expression of at least one MF-eczema differential gene, wherein the measurement of the fold change includes measuring the order of magnitude change in log2 FoldChange of at least one MF-eczema differential gene. The system is set to Identify at least one of the MF-eczema differential genes, and The at least one MF-eczema differential gene is identified as At least one of the at least MF-eczema differential genes associated with eczema, and At least one of the at least MF-eczema differential genes is associated with cutaneous lymphoma. The system is configured to identify at least one MF-eczema identification gene based on the skin condition threshold.

15. The system of any one of claims 9 to 14, wherein the system comprises at least one machine learning module, wherein the system comprises a training module configured to train the at least one machine learning module using any data in any step of the method, wherein the training is configured to train the at least one machine learning module according to any step of any of the preceding method claims, wherein the normalization method comprises performing M-value trimmed mean (TMM) normalization, wherein the system is configured to identify at least one sample dataset to predict the skin condition state.

16. The system according to any one of claims 9 to 15, wherein the system is configured to Identify at least one stage of the mycosis fungoides; and Distinguish between the at least one stage. The at least one stage includes at least one of the following: erythematous stage, plaque stage, tumor stage, and erythrodermic stage.

17. A kit for diagnosing eczema or mycosis fungoides, the kit comprising at least one means for quantifying the expression of at least one biomarker in at least one sample, wherein the method is the method according to any of the preceding method claims.

18. A kit for use in a method for differentiating mycosis fungoides from eczema or psoriasis, the kit comprising at least one means for quantifying the expression of at least one biomarker in at least one sample, wherein the method is the method according to any of the preceding method claims.

Images