A machine learning system for estimating thyroid profile from correlated clinical data

Abstract

A machine learning system for estimating thyroid profile parameters, consisting of: a data entry module configured to receive triiodothyronine concentration values ​​(T3) from clinical test results; a memory configured to temporarily store the input data; a data preprocessing unit configured to normalize and validate the received T3 concentration values, wherein preprocessed data is also stored in the memory; a machine learning module configured to estimate the TSH and T4 values ​​based on the input value of T3, comprising: a regression processor configured to estimate thyroid-stimulating hormone (TSH) levels from the preprocessed T3 concentration values ​​using the Ridge polynomial regression model; and a linear regression processor configured to estimate thyroxine (T4) levels from the preprocessed T3 concentration values ​​using the Gradient Boost Linear Regression model; a correlation analysis unit configured to validate the estimated TSH and T4 values ​​based on predefined correlation coefficients; a prediction engine configured to generate final thyroid profile parameters that include the estimated TSH and T4 values; and an output data module operatively connected to the prediction engine and configured to display the final thyroid profile parameters, including estimated TSH and T4 values, via a user interface display.

Description

FIELD OF THE INVENTIONThe present disclosure relates to a machine learning system for estimating thyroid profile from correlated clinical data. More particularly, the present invention relates to a machine learning system configured to estimate thyroid stimulating hormone (TSH) and thyroxine (T4) values from triiodothyronine concentrations (T3) using a machine learning based gradient boost regression model.BACKGROUND OF THE INVENTIONPrevious health predictive analysis attempts have focused primarily on epidemiological trends and material consumption using traditional statistical methods. Although these approaches have met with some success, they have had significant restrictions on accuracy and computing power. For example, while the system for predicting influenza trends based on Google trends data suffered from precision problems and an excessively simplified linear modeling. Also, although the aero material consumption prediction model was comprehensive, it required large computational resources and was sensitive to the selection of the input variables.The present invention overcomes these limitations by an innovative approach to determining thyroid profile using advanced machine learning techniques. Unlike prior techniques which rely heavily on computing resources or for simplicity, do not require precision, this invention utilizes gradient boost regression to achieve highly accurate estimates of thyroid parameters with consistent computing power. The system is particularly focused on determining the levels of thyroid stimulating hormone (TSH) and thyroxine (T4) from triiodothyronine concentrations (T3).This invention not only overcomes the computational and accuracy constraints of prior systems, but also provides a cost effective solution for thyroid profile analysis, making it particularly valuable for economically compromised populations. By reducing the cost of thyroid profiling to about one third of the cost of conventional methods while maintaining high accuracy, the invention represents a significant advance in available health diagnostics.SUMMARY OF THE INVENTIONThe present disclosure relates to a machine learning system for estimating thyroid profile from correlated clinical data. The invention provides a novel system for cost effective estimation of thyroid profile using advanced machine learning techniques. The system uses gradient boost regression and correlation analysis to estimate TSH and T4 values from T3 concentrations. This significantly reduces the test costs and at the same time ensures high accuracy. This solution facilitates creating a comprehensive thyroid profile to economically impaired population groups.The present disclosure aims to provide a machine learning system for estimating thyroid profile from correlated clinical data. The system comprises: a data input module for receiving triiodothyronine (T3) concentration values from clinical test results; a memory for temporarily storing the input data; a data preprocessing unit for normalizing and validating the received T3 concentration values, wherein preprocessed data is also stored in the memory; a machine learning module for estimating the TSH and T4 values based on the T3 input value, consisting of a regression processor for estimating the TSH (thyroid stimulating hormone) values from the preprocessed T3 concentration values using the ridge polynomial regression model; A linear regression processor for estimating thyroxine (T4) values from the preprocessed T3 concentration values using the gradient boost linear regression model; a correlation analyzer for validating the estimated TSH and T4 values from predefined correlation coefficients; a prediction engine for generating final thyroid profile parameters comprising the estimated TSH and T4 values; and an output data module operatively connected to the prediction engine and configured to display the final thyroid profile parameters comprising the estimated TSH and T4 values via a user interface display.An object of the present disclosure is to provide a machine learning system for estimating thyroid profile from correlated clinical data.Another object of the present disclosure is to develop a low cost system for estimating total thyroid profiles by reducing the number of laboratory tests required.Another object of the present disclosure is to achieve highly accurate estimation of TSH and T4 values using only the T3 concentration values using advanced machine learning models.Another object of the present disclosure is to make comprehensive thyroid studies more accessible to economically impaired populations while maintaining clinical reliability.In order to further clarify the advantages and features of the present disclosure, the invention will be explained in more detail with reference to specific embodiments that are illustrated in the accompanying drawings. These drawings illustrate only typical embodiments of the invention and are therefore not to be considered as limiting the scope thereof. The invention will be described and explained in more detail with reference to the accompanying drawings.BRIEF DESCRIPTION OF THE FIGURESThese and other features, aspects, and advantages of the present disclosure will become more fully understood when the following detailed description is read with reference to the accompanying drawings, in which like characters represent like parts throughout. The following applies here: FIG. 1 is a block diagram of a machine learning system for estimating thyroid profile from correlated clinical data, according to an embodiment of the present disclosure; and FIG. 2 illustrates a diagram showing the operation of the proposed system according to an embodiment of the present disclosure.Those skilled in the art will also appreciate that the elements in the drawings are shown for simplicity and are not necessarily to scale. For example, the flowcharts illustrate the method using the key steps to improve understanding of aspects of the present disclosure. In addition, regarding the construction of the apparatus, individual or multiple components of the apparatus may be represented by conventional symbols in the drawings. The drawings may only show the specific details relevant to understanding the embodiments of the present disclosure in order not to obscure the drawings with details readily apparent to those skilled in the art after the present description.DETAILED DESCRIPTION:In order to aid in the understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and will be described in an comprehensible manner. However, the scope of the invention is not limited thereby. Changes and further modifications of the illustrated system, as well as further applications of the principles of the invention, are possible, as would normally occur to a person skilled in the art.It will be understood by those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be limiting thereof.References throughout this specification to "one aspect," "another aspect," or similar language mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, the phrases "in one embodiment," "in another embodiment," and similar phrases in this specification may or may not refer to the same embodiment.The terms "comprises," "comprising," or variations thereof cover a nonexclusive inclusion. A process or method comprising a list of steps not only includes these steps, but may also include additional steps not expressly listed or inherent to the process or method. Likewise, the phrase "comprises... for" one or more devices, subsystems, elements, structures, or components does not exclude, without further limitations, the existence of further devices, subsystems, elements, structures, or components, or additional devices, subsystems, elements, structures, or components.Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by one of ordinary skill in the art. The systems, methods, and examples provided herein are for illustrative purposes only and are not to be considered limiting.Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.The functional units described in this specification are referred to as devices. A device may be implemented in programmable hardware devices such as processors, digital signal processors, central processing units, field programmable gate arrays, programmable array logic systems, programmable logic devices, cloud processing systems, or the like. The devices may also be implemented in software for execution by various types of processors. An identified device may include executable code and may consist, for example, of one or more physical or logical blocks of computer instructions, which may be organized, for example, as an object, procedure, function, or other construct. However, the executable of an identified device need not be physically stored at the same location, but may consist of different instructions stored at different locations that, logically linked, form the device and serve its purpose.Device or module executable code may consist of one or more instructions and even be distributed over multiple code segments, different applications, and multiple storage devices. Likewise, operational data may be identified and displayed within the device and organized in any form and data structure. The operational data may be acquired as a single data set or distributed across different storage devices and may be at least partially present as electronic signals in a system or network.References throughout this specification to "a selected embodiment," "an embodiment," or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosed subject matter. Therefore, the terms "a selected embodiment," "in one embodiment," or "in one embodiment" in various places throughout this specification do not necessarily refer to the same embodiment.Moreover, the described features, structures, or characteristics may be combined in any manner in one or more embodiments. The following description contains numerous specific details to provide a thorough understanding of the embodiments of the disclosed subject matter. However, those skilled in the art will appreciate that the disclosed subject matter may be practiced without one or more of the specific details or with other methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail in order not to obscure aspects of the disclosed subject matter.According to the example embodiments, the disclosed computer programs or modules may be executed in a variety of ways, such as an application in a device's memory or a hosted application on a server that communicates with the device application or browser via various standard protocols such as TCP / IP, HTTP, XML, SOAP, REST, JSON, and other suitable protocols. The disclosed computer programs may be written in example programming languages that execute from the memory of the device or from a hosted server, such as BASIC, COBOL, C, C++, Java, Pascal, or scripting languages such as JavaScript, Python, Ruby, PHP, Perl, or other suitable programming languages.Some of the disclosed embodiments include or otherwise involve data transfer over a network, for example, the transfer of various inputs or files over the network. The network may include, for example, the Internet, wide area networks (WANs), local area networks (LANs), analog or digital wired and wireless telephone networks (e.g., PSTN, Integrated Services Digital Network (ISDN), cellular networks and digital subscriber line (xDS)), radio, television, cable, satellite, and / or other transmission or tunneling mechanisms for data transmission. The network may comprise multiple networks or sub-networks, each including, for example, a wired or wireless data path. The network may comprise a circuit switched voice network, a packet switched data network or other network for transferring electronic communication. For example, the network may comprise networks based on Internet Protocol (IP) or Asynchronous Transfer Mode (ATM) and support voice, for example, via VoIP, voice over ATM, or other comparable protocols for voice data communication. In one implementation, the network includes a cellular network configured to exchange text or SMS messages.Examples of the network include a personal area network (PAN), a storage area network (SAN), a home area network (HAN), a campus area network (CAN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a virtual private network (VPN), an enterprise private network (EPN), the Internet, a global area network (GAN), etc.FIG. 1 shows a block diagram of a machine learning system ( 100) for estimating thyroid profile from correlated clinical data according to an embodiment of the present disclosure.Referring to FIG. 1, the machine learning system (100) includes: a data input module (102) configured to receive triiodothyronine concentration values (T3) from clinical test results; a memory (104) configured to temporarily store the input data; a data preprocessing unit (106) configured to normalize and validate the received T3concentration values, wherein preprocessed data is also stored in the memory (104); a machine learning module (108) configured to estimate the TSH and T4 values based on the T3 input value, comprising: a regression processor configured to estimate the TSH (thyroid stimulating hormone) values from the preprocessed T3 concentration values using a ridge polynomial regression model; and a linear regression processor configured to estimate the thyroxine (T4) values from the preprocessed T3 concentration values using a gradient boost linear regression model; a correlation analysis unit (110) configured to validate the estimated TSH and T4 values based on predefined correlation coefficients; a prediction engine (112) configured to generate final thyroid profile parameters comprising the estimated TSH and T4 values; and an output data module (114) operatively connected to the prediction engine (112) and configured to display the final thyroid profile parameters comprising the estimated TSH and T4 values via a user interface display (116).In one embodiment, the data preprocessing unit (106) is also configured to perform feature extraction for the received T3concentration values, remove outliers from the received data, and standardize the data format for compatibility with machine learning models.In one embodiment, the regression processor (108a) and the linear regression processor (108b) are trained using a dataset of at least 2500 clinical recordings containing known T3, T4, and TSH values to obtain a first correlation coefficient for TSH estimation and a second correlation coefficient for T4 estimation.In an embodiment, the correlation analysis unit (110) comprises a heat map generator (110a) configured to visualize correlations between T3, T4and TSHvalues, and a validation module (110b) configured to check whether the correlation coefficient of the TSH estimate is 95.83% and the correlation coefficient of the T4estimate is 92.41%.In one embodiment, the system (100) further includes a confidence assessment module (118) configured to assess the reliability of the estimates generated, flagging estimates that are outside predetermined confidence intervals, and trigger additional validation for flagged estimates.In one embodiment, prediction engine (112) also includes a boundary case detection module (112a) configured to identify boundary cases in the estimated thyroid profile parameters, classify detected boundary cases as hyper or hypo states, and generate clinical correlation alerts when boundary cases are detected.In one embodiment, the system (100) further comprises a data storage unit (120) configured to manage a database of historical thyroid profile parameters, store correlation patterns between T3, T4 and TSH values, and update correlation patterns based on new validated data.In one embodiment, the machine learning module ( 108) is also configured to retraining the regression models periodically using newly validated clinical data, optimize the model parameters based on metrics for prediction accuracy, and maintain version control of the trained models.In one embodiment, the system (100) further comprises a reporting module (122) operatively connected to the prediction engine (112) and configured to generate detailed reports of estimated thyroid profile parameters, include confidence scores for each estimation, and provide comparative analysis with standard reference ranges.In one embodiment, the system (100) further comprises an integration interface (124) in communication with a user interface display (116) configured to: connect to laboratory information management systems, automatically receive T3 test results, and export estimated thyroid profile parameters in standardized formats.In one embodiment, the proposed system is limited to the normal range of T3concentration. The determination of T4 and TSH is possible when the T3 value is in the normal range.The present invention relates to a machine learning based system that receives T3concentration values, processes them by trained machine learning models, and generates accurate estimates of the TSH and T4values. The system uses the ridge polynomial regression model for TSH estimation and the gradient boost linear regression model for T4 estimation and integrates validation mechanisms to ensure clinical reliability. The system includes components for data preprocessing, model training, correlation analysis and result validation to ensure accurate estimation of the thyroid profile.FIG. 2 illustrates a diagram showing the operation of the proposed system according to an embodiment of the present disclosure.Figure 2 shows that the system is configured for the analysis of new test parameters in the clinical or diagnostic context. It uses the T3parameter value as input to determine the TSH and T4values. The system is configured to process a new test parameter (T3parameter) as input and determine relevant relationships and contexts along with correlated parameters. The system then takes into account available attributes representing supporting data or characteristics of the parameter to allow for more detailed analysis.The core of the system is a database with known cases and established results. The system compares the new parameter and its attributes to this database, using previous data to identify patterns or similarities. By this comparison, the system can judge whether the parameter coincides with existing cases or needs more accurate examination. If the analysis yields a marginal result, the system can determine whether the parameter has hypo (low concentration) or hyper (high concentration) tendencies. If the result is marginal, the system identifies a clinical correlation that requires additional validation by clinical data or other testing.After the comparison is completed, the system attempts to diagnose the parameter based on the confidence level determined from the analysis. If the system determines that the confidence level is sufficient, it predicts a result and provides an output reflecting its score. However, if the confidence level is insufficient, the system refines its analysis by repeating the comparison with the pool of known cases, thus ensuring a more accurate assessment prior to result generation.The system is designed for accuracy and reliability, in particular with limit-value or non-clear results. By integrating reference data into systematic analyses and allowing clinical validations when needed, the system ensures robust and reliable results in diagnostic or predictive scenarios. The iterative character allows continuous refinement and focuses on the precision in handling test parameters.The invention focuses on parameters derived from the immunoassay segment for experiments and aims to provide a cost effective alternative for estimating thyroxine (T4) and TSH (thyroid stimulating hormone (TSH) values from triiodothyronine (T3) concentrations. Laboratory tested values of T3concentrations were used as inputs to interpolated equations, while the concentrations of T4and TSHwere predicted during the test phase. This approach not only reduces test costs, but also ensures accessibility in areas where such diagnostic profiles are financially not feasible for low-revenue population groups. The object of the invention is to investigate the correlation between parameters in different areas of pathology, such as immunoassay, biochemistry and hematology. A generative AI-based machine learning model was trained and validated thoroughly using a dataset with concentration values of TSH, T3, and T4. Ridge polynomial regression and gradient boost linear regression were used for prediction. This allows the model to predict precisely the TSH and T4 concentrations in the test phase. This invention reduces the overall cost of the tests by up to one third and offers considerable potential for use in economically disadvantageous regions, thus ensuring cost effective and reliable diagnostic solutions.In one embodiment, the machine learning system uses regression models to estimate the TSH and T4 values, training these models from a 2500 data dataset containing T3, T4, and TSH concentration values of different patients that have been trained and validated, with the data used for training being preprocessed.The system uses various machine learning models to identify accurate correlations between the parameters TSH, T3, and T4. The data set is feature extracted to isolate and refine key attributes. This allows the system to focus on the most relevant information for the analysis. Correlation heat maps are generated to visually represent the relationships between the components and highlight the strongest correlations between any two parameters. Testing of various approaches has determined that T4can be estimated from T3with high accuracy using linear regression. In addition, T3 serves as a reliable predictor for the estimation of TSH values. This estimation is achieved by employing polynomial regression and gradient boosting regression models which provide robust and accurate predictions. The invention relies on an iterative and data controlled approach to ensure accurate modeling and parameter estimation. It is important to note that the proposed system works only for the normal range of T3concentration. The estimation of T4 and TSH can be made only when the test results show a T3 concentration in the normal range.The regression coefficient shows a strong relationship between T3-TSH and T3-T4 and emphasizes the effectiveness of the model in acquiring these relationships. For T3-TSH, the model achieves an accuracy of R 2= 95.83 % at a mean square error (MSE) of 1.12; for T3-T4, the accuracy is also R 2= 92.41 % at an MSE of 0.98. The minimum absolute error is 0.58 and indicates high precision. This error can be further minimized by training the regression model with a larger dataset, which would also improve the overall accuracy of the system.Clinical diagnoses are indeed decisive for the start of treatment, but are often prohibitively expensive and therefore inaccessible to low-revenue groups. The proposed model addresses this challenge by significantly reducing diagnostic costs and thereby making them more affordable and accessible. By concentrating on detected anomalies, the model optimizes the diagnostic process and thus saves time and resources with consistent accuracy and reliability.The drawings and the foregoing description show examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be divided into multiple functional elements. Elements of one embodiment may be added to another embodiment. For example, the order of the processes described herein may be changed and is not limited to the manner described herein. Moreover, the actions of a flow chart need not be performed in the order shown; nor do all actions necessarily need to be performed. Also, actions that are not dependent on other actions may be performed in parallel with the other actions. The scope of the embodiments is by no means limited by these specific examples. Numerous variations, whether or not explicitly stated in the specification, such as differences in structure, dimensions, and material use, are possible. The scope of the embodiments is at least as broad as recited in the following claims.Advantages, other advantages and solutions to problems have been described above with reference to specific embodiments. However, the advantages, merits, solutions to problems and any components that may result in an advantage, benefit or solution being introduced or enhanced are not to be understood as critical, required or essential features or components of individual or all claims.REFERENCES100 A machine learning system for estimating thyroid profile from correlated clinical data. 102 Data input module 104 Memory 106 Data preprocessing unit 108 Machine learning module 108 a Regressions processor 108 bLinear regression processor 110 Correlation analysis unit 112 Prediction engine 112 a Grenzlinie recognition module 114 Output data module 116 Display of the user interface 118 Module for confidence assessment 120 Data storage unit 122 Reporting module 124 Integration interface 202 Pool of known cases 204 New test parameter 206 Correlated parameter 208 Available attributes 210 Comparison with the pool of known cases 212 Checking for whether the result is marginal 214 Checking for the hypo- or hyperconcentration of the parameter 216 Clinical correlation required 218 Attempt for diagnosis 220 Results predicted 222 Enough self confidence 224 Not enough confidence

Claims

A machine learning system for estimating thyroid profile parameters, comprising: a data input module configured to receive triiodothyronine concentration values (T3) from clinical test results; a memory configured to temporarily store the input data; a data preprocessing unit configured to normalize and validate the received T3 concentration values, wherein preprocessed data is also stored in the memory; a machine learning module configured to estimate the TSH and T4 values based on the input value of T3, comprising: a regression processor configured to estimate thyroid stimulating hormone (TSH) values from the preprocessed T3 concentration values using the ridge polynomial regression model; and a linear regression processor configured to estimate thyroxine values (T4) from the preprocessed T3 concentration values using the gradient boost linear regression model; a correlation analysis unit configured to validate the estimated TSH and T4 values based on predefined correlation coefficients; a prediction engine configured to generate final thyroid profile parameters comprising the estimated TSH and T4 values; and an output data module operatively connected to the prediction engine and configured to display the final thyroid profile parameters including the estimated TSH and T4 values via a user interface display.The system of claim 1, wherein the data preprocessing unit is further configured to perform feature extraction for the received T3 concentration values, remove outliers from the received data, and standardize the data format for compatibility with machine learning models.The system of claim 1, wherein the gradient boost regression process and the linear regression processor are trained using a dataset of at least 2500 clinical recordings containing known T3, T4, and TSH values to achieve a first correlation coefficient for TSH estimation and a second correlation coefficient for T4 estimation.The system of claim 1, wherein the correlation analysis unit comprises a heat map generator configured to visualize correlations between T3, T4, and TSH values; and a validation module configured to verify whether the correlation coefficient of the TSH estimate is 95.83%; and the correlation coefficient of the T4 estimate is 92.41%.The system of claim 1, further comprising a confidence assessment module configured to assess the reliability of the estimates generated, mark estimates that are outside predetermined confidence intervals, and trigger additional validation for marked estimates.The system of claim 1, wherein the prediction engine further comprises a boundary case detection module configured to: identify boundary cases in the estimated thyroid profile parameters, classify detected boundary cases as hyper- or hypo-states, and generate clinical correlation alerts when boundary cases are detected.The system of claim 1, further comprising a data storage unit configured to manage a database of historical thyroid profile parameters, to store correlation patterns between T3, T4 and TSH values, and to update correlation patterns based on new validated data.The system of claim 1, wherein the machine learning module is further configured to retraining the regression models periodically using newly validated clinical data, optimize the model parameters based on metrics for prediction accuracy, and maintain version control of the trained models.The system of claim 1, further comprising a reporting module operatively connected to a prediction engine and configured to generate detailed reports about the estimated thyroid profile parameters, include confidence values for each estimate, and provide comparative analysis with standard reference ranges.The system of claim 1, further comprising an integration interface in communication with a user interface display configured to: communicate with laboratory information management systems, automatically receive T3 test results, and export estimated thyroid profile parameters in standardized formats.

Images