Biological image transformation using machine-learning models

Abstract

83 20 25 28 74 03 30 D ec 2 02 5 A B S T R A C T 2 0 2 5 2 8 7 4 0 3 3 0 D e c 2 0 2 5 8 3 IL LU M IN AT IO N PA TT ER N (S ) 11 0 A SE CO ND T YP E OF IM AG ES (E .G ., FL UO RE SC EN CE IM AG ES ) 12 4 A FI RS T TY PE O F IM AG ES (E .G ., BR IG HT -F IE LD IM AG ES 12 2 DI SC RI MI NA TO R 10 4 GE NE RA TO R 10 2 Tr ain in g Da ta 12 0 GA N Mo de l 1 00 FI G. 1 1 / 26 Training Data 120 GAN Model 100 ILLUMINATION A FIRST TYPE OF IMAGES (E.G., PATTERN(S) BRIGHT-FIELD IMAGES 110 122 GENERATOR 102 A SECOND TYPE OF IMAGES (E.G., DISCRIMINATOR FLUORESCENCE IMAGES) 104 124 FIG. 1 20 25 28 74 03 30 D ec 2 02 5 2 0 2 5 2 8 7 4 0 3 3 0 D e c 2 0 2 5 T r a i n i n g D a t a 1 2 0 ILLUMINATION A FIRST TYPE OF IMAGES (E.G., P A T T E R N ( S ) BRIGHT-FIELD IMAGES A SECOND TYPE OF IMAGES (E.G., FLUORESCENCE IMAGES)

Description

[0340] For example, at block 1106, the system trains the model using images corresponding to a first set of SLM configurations. Each image results in a corresponding loss based on a loss function of the model. At block 1108, the system determines which SLM configuration of the first set of SLM configurations results in the smallest loss (e.g., generator loss). At block 1110, the SLM configuration that has produced the smallest loss can be identified and a new second set of SLM configurations can be identified accordingly. For example, the new set of SLM configurations can include the best SLM configuration (i.e., the configuration that produced the smallest loss) from the first set and / or one or more new SLM configurations similar to the best SLM configuration. The new set of SLM configurations can also exclude SLM configurations that have resulted in the biggest loss from the first set. The new set of SLM configurations can be loaded onto the optical system to obtain additional training data. This steps can be repeated until a threshold is met, for example, when no more improvement (e.g., on the generator loss) is observed. The optimal SLM configuration can be stored and used to obtain input images.

[0341] While steps 1106-1112 are described as part of a training process, they can be performed in other stages of the pipeline (e.g., inference stage) to identify the optimal SLM configuration for generating input images. In some embodiments, the light source and the SLM of the optical system can be iteratively programmed together to identify the best combination of illumination pattern and SLM configuration for generating input images.

[0342] While FIGS. 9-11 describes optimization techniques using an SLM component of an optical system, it should be appreciated the SLM can be replaced with another hardware component that can alter the optical function (e.g., pupil function) of the system, such as micromirrors, without departing from the spirit of the invention.

[0343] FIGS. 12A and 12B illustrate a side-by-side comparison of classification results of two classification models, in accordance with some embodiments. The classification model in FIG. 12A is trained using real images captured by a microscope (e.g., real fluorescence images), while the classification model in FIG. 12B is trained using synthetic images generated using the techniques described herein (e.g., fluorescence images generated from bright-field images). The classification models determines which chemical compound that tissues depicted in an input image have responded to. Specifically, each model is configured to receive an input image and output a classification result indicative of one of 150 predefined chemical compounds. In the depicted example, each of FIGS. 12A and 12B shows a Uniform Manifold 2025287403   30 Dec 2025 Approximation and Projection (UMAP) in which each input image is represented as a point in the UMAP. The color of the point represents the chemical compound the image is classified as by the classification model. In some embodiments, the input images into the model in FIG. 12A are real images, while the input images into the model in FIG. 12B are generated images.

[0344] FIGS. 12C and 12D illustrate a side-by-side comparison of the same classification results in FIGS. 12A and 12B, respectively, with a different color scheme to demonstrate the two models’ resistance to batch effects. Batch effects refer to situations where subsets (i.e., batches) of data significantly differ in distribution due to irrelevant, instrument-related factors. Batch effects are undesirable because they introduce systematic errors, which may cause downstream statistical analysis to produce spurious results and / or obfuscate the signal of interest. In each of FIGS. 12C and 12D, the input images belong to three different batches (e.g., from different plates or experiments), as indicated by different grey levels. As shown, FIG. 12D shows a greater overlap among the points corresponding to the three batches, indicating that the generated images are more resistance to batch effects. FIG. 12E illustrates the Euclidean distance metric corresponding to a real image set (e.g., input images as shown in FIGS. 12A and 12C), a generated image set (e.g., input images as shown in FIGS. 12B and 12D), and a truly batch-invariant image set. The Euclidean distance metric for each image set measures the Euclidean distance between the image embeddings from two separate batches (e.g., Real Image Batch 1 and Real Image Batch 3 in FIG. 12C, Generated Image Batch 1 and Generated Image Batch 3 in FIG. 12D). For a truly batch-invariant image set, the mean score should be 0 (i.e., no batch effects). As shown, the mean score for the generated image set is lower than the mean score for the real images, thus demonstrating superior batch invariance.

[0345] FIGS. 13A and 13B illustrate an exemplary generated phase image and an exemplary generated fluorescence image, in accordance with some embodiments. FIG. 13A is a phase image generated by a GAN model described herein, for example, from a bright-field image. FIG. 13B is a fluorescence / bodipy image generated by the GAN from the same bright-field image. The bodipy image includes a virtual green hue to highlight the presence of a biomarker. FIG. 13C shows FIG. 13B overlaid onto the phase image in FIG. 13A. Thus, the GAN model can be used for disease modeling. For example, multiple samples can be obtained and imaged from a subject. The resulting series of bright-field images can be analyzed by the GAN model to generate synthetic phase and fluorescence images to study disease perturbations.

[0346] FIGS. 14A-B illustrate an exemplary process for training a machine-learning model (e.g., a GAN model) configured to generate synthetic data (e.g., image data) and identifying an optimal illumination scheme for obtaining input data for the machine-learning model, in 2025287403   30 Dec 2025 accordance with some embodiments. Process 1400 is performed, for example, using one or more electronic devices implementing a software platform. In some examples, process 1400 is performed using a client-server system, and the blocks of process 1400 are divided up in any manner between the server and one or more client devices. In other examples, process 1400 is performed using only one or more client devices. In process 1400, some blocks are, optionally, combined, the order of some blocks is, optionally, changed, and some blocks are, optionally, omitted. In some examples, additional steps may be performed in combination with the process 1400. Accordingly, the operations as illustrated (and described in greater detail below) are exemplary by nature and, as such, should not be viewed as limiting.

[0347] At block 1402, an exemplary system (e.g., one or more electronic devices) receives a plurality of training images. The plurality of training images are real images of biological samples and are also referred to as the ground truth data. The plurality of training images comprises the type of image data that the GAN model is configured to receive and the type(s) of image data that the GAN model is configured to output. For example, if the GAN model is configured to receive a bright-field image and output a fluorescence image and a phase image, the received plurality of images would include a plurality of bright-field training images 1402a (i.e., a GAN input data type), a plurality of fluorescence training images 1402b (i.e., a GAN output data type), and a plurality of phase training images (i.e., a GAN output data type).

[0348] The bright-field training images in the plurality 1402a can be captured by illuminating in vitro (or biopsy) cell samples with an inexpensive LED array using different illumination settings. The fluorescence training images in the plurality 1402b can be captured after a dye is applied to the biological sample (e.g., to enhance the visibility of a biomarker). The phase training images can be obtained using physics- or optics-based models. It should be appreciated by one of ordinary skill in the art that the plurality of training images used in the process 1400 can differ depending on the type of the image data that the GAN is configured to receive and output. In some embodiments, the plurality of training images comprises paired image data. For example, a bright-field image, a fluorescence image, and a phase image of the same biological sample can be included in the sets 1402a, 1402b, and 1402c, respectively.

[0349] In some embodiments, the plurality of training images are acquired to enable training of the GAN model such that the synthetic images (e.g., synthetic fluorescence images, synthetic phase images) generated by the GAN model will provide the same performance in downstream analyses as real images (e.g. real fluorescence images, real phase images). In some embodiments, the downstream analyses include a classification task that classifies an image as corresponding to one class out of M classes. For example, the classification task can involve 2025287403   30 Dec 2025 classifying an image as corresponding to a particular cell state out of multiple cell state classes (e.g., healthy state, diseased state). As another example, the classification task can involve classifying an image as corresponding to a particular perturbation out of multiple perturbation classes. In order to train the GAN model to generate synthetic images that can be classified as accurately as the real images, the training images include images corresponding to the M classes (or conditions). For example, if the M classes include a healthy cell state class and a diseased cell state class, the plurality of bright-field training images 1402a can include bright-field images depicting healthy cells and bright-field images depicting diseased cells, the plurality of fluorescence training images 1402b can include fluorescence images depicting healthy cells and fluorescence images depicting diseased cells, and the plurality of phase training images 1402c can include phase images depicting healthy cells and phase images depicting diseased cells. For example, if the M classes include M perturbations, the plurality of bright-field training images 1402a can include bright-field images depicting the M perturbations, etc. Each training image can be labelled with the corresponding condition. For example, a phase image depicting a diseased cell state can be associated with a diseased label.

[0350] In an exemplary implementation, the plurality of training images includes X fields of view per condition. i.e., M x X fields of view in total. Specifically, in the plurality of bright-field images 1402a, each field of view includes N bright-field images captured using N illumination settings, thus resulting in M x X x N bright-field images in total. In the plurality of fluorescence images 1402b, each field of view includes one fluoresce image, thus resulting in M x X fluorescence images in total. In the plurality of phase images 1402c, each field of view includes one phase image, thus resulting in M x X phase images in total. In some embodiments, the bright-field images are at magnification = min, while the fluorescence images are at magnification mout >= min and the phase images are at mout >= min.

[0351] At block 1404, the system trains a classifier configured to receive an input image and output a classification result indicative of one of M conditions. For example, if the M conditions include a healthy condition and a diseased condition, the classifier is configured to receive an input image and output a classification result indicative of either the healthy condition or the diseased condition. After the classifier is trained, it is used during the training of the GAN model to ensure that the GAN model can generate synthetic image data that can be classified to the same or similar level of accuracy as real image data, as described below.

[0352] In some embodiments, the classifier is trained using the same type of image data that the GAN model is configured to output. In the depicted example in FIGS. 14A and 14B, the GAN model is configured to output a fluorescence image and a phase image; thus, the classifier 2025287403   30 Dec 2025 trained in block 1404 is trained using the plurality of fluorescence training images 1402b and the plurality of phase training images 1402c. During training, each fluorescence image or phase image is inputted into the classifier to obtain a predicted classification result (e.g., healthy or diseased). The predicted classification result is then compared against the actual class associated with the training image (e.g., whether training image in fact depicts a healthy cell state or diseased cell state) and, based on the comparison, the classifier can be updated accordingly. The classifier can be implemented using any classification algorithm, such as a logistic regression model, a naive Bayes model, a decision tree model, a random forest model, a support vector machine model, etc.

[0353] At block 1406, the system trains the GAN model based on the training images. Block 1406 can include steps 1408a-1408e, which can be repeated until the training is complete (e.g., when convergence is reached). The steps 1408a-e are described below with reference to FIG. 15, which is a schematic diagram illustrating the steps, in accordance with some embodiments.

[0354] As shown in FIG. 15, the GAN model includes a multi-head attention layer 1502 comprising a matrix of weights, specifically, K sets of weights w1-wn, a generator 1504, a discriminator 1508, and the trained classifier 1506 from block 1404. During training of the GAN model, the classifier remains fixed while the generator, the discriminator, and the attention layer are updated, as described below.

[0355] At block 1408a, the system applies each of K sets of weights in the attention layer of the GAN model to a set of bright-field training images. The set of bright-field training images are obtained from the plurality of bright-field images 1402a. In some embodiments, the set of bright-field training images correspond to the same field of view and depict the same biological sample, but are captured using different illumination settings. For example, if a LED array comprises N illumination emitters (e.g., LEDs), each illumination emitter can be turned on one at a time and a bright-field image of the biological sample illuminated by each illumination emitter can be captured, thus resulting in a set of N bright-field training images.

[0356] In the depicted example in FIG. 15, the system receives a set of N bright-field training images corresponding to illumination settings 1-N. For example, the first bright-field image depicts the biological sample being illuminated using the illumination setting 1 (e.g., only the first LED in the array is turned on), the second bright-field image depicts the biological sample being illuminated using the illumination setting 2 (e.g., only the second LED in the array is turned on), . . . and the N-th bright-field image depicts the biological sample being illuminated using the illumination setting N (e.g., only the N-th LED in the array is turned on). 2025287403   30 Dec 2025

[0357] The attention layer generates K sets of weights and each set comprises N weights. Each set of weights w1-wn is applied to the N images to generate an aggregated image. For each set of weights, the attention layer 1502 assigns a continuous weight (e.g., a normalized scalar weight) in the set to each of the set of bright-field training images. These weights correspond to intensity values of the corresponding illumination settings (e.g., the corresponding LEDs). As shown, w1 is applied to (e.g., multiplied with) the first bright-field image, w2 is applied to the second bright-field image, wn is applied to the N-th bright-field image. After the weights are applied, the weighted images can be aggregated (e.g., summed) to obtain one aggregated bright-field image. Because there are K sets of weights, K aggregated images 1512 can be generated. In some embodiments, the attention layer is an adapted multi-head attention layer. The attention mechanism allows the natural generation of K linear combination of bright-field images (i.e., aggregated images). These aggregated images are fed to the rest of the network as described herein.

[0358] At block 1408b, the system inputs each of the aggregated bright-field images into the GAN model. With reference to FIG. 15, each of the aggregated bright-field images 1512 is inputted into the generator 1504, which outputs a synthetic fluorescence image 1514a and a synthetic phase image 1514b. The generator can be implemented in a similar manner as the generator described above with reference to FIGS. 3A-D.

[0359] During training, the generator output (i.e., the generated fluorescence image 1514a and the generated phase image 1514b ) can be connected directly to the discriminator input. The discriminator 1508 is trained to distinguish generated images and real images. During training, a generated image can be inputted into the discriminator 1508 to obtain a discriminator loss and a generator loss as described above. Further, a real image can also be inputted into the discriminator 1508 to generate a discriminator loss and a generator loss. The real image can be the real fluorescence image 1516a (from the plurality of fluorescence training images 1402b in FIG. 14A) corresponding to the same field of view as the input bright-field images, or the real phase image 1516b (from the plurality of phase training images 1402c in FIG. 14A) corresponding to the same field of view as the input bright-field images.

[0360] In some embodiments, the discriminator loss function is a Wasserstein discriminator loss and calculated as follows: ** 1a£ ^)- / (^°)) i=i 2025287403   30 Dec 2025

[0361] where f(x) is the discriminator’s output based on wavelet coefficients of a real fluorescence or phase image, w is the model weights of the discriminator, m is the size of the mini-batch, f is the discriminator model, x is the real image, z is the input (bright-field image 1512), G is the generator model, and f(G(z)) is the discriminator’s output based on the predicted wavelet coefficients corresponding to a synthetic fluorescence or phase image.

[0362] In some embodiments, the generator loss function is a Wasserstein generator loss and calculated as follows: m 79;1 2    / -- - :i i=i

[0363] where f(x) is the discriminator’s output based on wavelet coefficients of a real fluorescence or phase image, m is the size of the mini-batch, f is the discriminator model, z is the input (bright-field image 1512), G is the generator model, and f(G(z)) is the discriminator’s output based on the predicted wavelet coefficients.

[0364] At block 1408c, the system inputs each generated image and a real image corresponding to the generated image into the trained classifier to obtain a classifier loss. For example, the generated fluorescence image 1514a is inputted into the classifier 1506 to obtain a first classification result; the real fluorescence image 1516a is inputted into the classifier 1506 to obtain a second classification result; a classifier loss can be calculated based on the difference between the first classification result and the second classification result. As another example, the generated phase image 1514b is inputted into the classifier 1506 to obtain a third classification result; the real phase image 1516b is inputted into the classifier 1506 to obtain a fourth classification result; a classifier loss can be calculated based on the difference between the third classification result and the fourth classification result.

[0365] At block 1408d, the system augments the generator loss based on the classifier loss. For example, the generator loss can be augmented with a L2 norm of the classification score from the real images. In some embodiments, if no classifier is available or no classification is wanted, the classifier loss is replaced by a constant (e.g. 0).

[0366] At block 1408e, the system updates the GAN model based on the augmented generator loss. The backpropagation follows the same procedure as described above. The discriminator updates its weights through back-propagation based on the discriminator loss through the discriminator network. Further, the augmented generator loss is back-propagated to update the weights in the attention layer (e.g., the weights with which the aggregated image is calculated) and the generator. For example, the generated loss calculated based on an aggregated image 2025287403   30 Dec 2025 corresponding to the K-th set of weights can be used to update the K-th set of weights in the attention layer.

[0367] At block 1410, the system obtains one or more optimal illumination patterns based on the weights in the attention layer of the trained GAN model. As described above, the weights in the attention layer (e.g., w1 - wn) can be indicative of the intensity values of the corresponding illumination settings (e.g., the corresponding LEDs in the LED array). As discussed above, the attention layer can be a multi-head attention layer that provide K sets of weights (i.e., K linear combinations), thus resulting in K illumination patterns.

[0368] In some embodiments, before the block 1406, the generator and the discriminator of the GAN model are pre-trained using bright-field images in which all LEDs in the LED array are turned on. After the pre-training, block 1406 is performed to update the attention weights while the generator and the discriminator remain fixed. The optimal combination of the illuminations can be obtained based on the updated weights.

[0369] FIG. 16A illustrates synthetic images generated by an exemplary GAN model to study the NASH disease, in accordance with some embodiments. In the depicted example, Hepg2 cells with NASH genetic background are illuminated using an optimal illumination pattern identified using techniques described herein, and a bright-field image is captured. The bright-field image is inputted into a GAN model, which outputs a generated phase image and a generated fluorescence / bodipy image in FIG. 16A. The GAN model used in FIG. 16A can be configured to include a trained classifier as described with reference to FIGS. 14A-B and 15. The classifier has two conditions: a healthy condition (e.g., no perturbations) and a diseased condition (e.g., inflammation cocktail plus fatty acid). As described, the GAN model can be trained such that the generated images can be classified by the classifier to the same or similar degree of accuracy as real images.

[0370] The synthetic images generated by the GAN model can be used to create disease models. FIG. 16B illustrates downstream analyses of the generated images, in accordance with some embodiments. As shown, the generated images can be used to perform nucleus segmentation and blob detection (e.g., using image-processing algorithms). Thus, the generated images can be used to create a disease model for the NASH disease and study chemical perturbations in NASH.

[0371] The synthetic images generated by the GAN model can also be used to evaluate the efficacy of a treatment. The system processes three groups of generated images: a first group of images depicting healthy tissues that do not have the disease, a second group of images depicting untreated diseased tissues, and a third group of images depicting diseased tissues that 2025287403   30 Dec 2025 have been treated (e.g., using a particular drug). In the depicted example in FIG. 16C, the first group of images (labelled “untreated”) comprises images of healthy tissues that do not have the non-alcoholic steatohepatitis (NASH) disease, the second group of images (labelled “NASH 2X”) comprises images of untreated tissues having the NASH disease, and the third group of images (labelled “NASH 2X+ACC inhibitor”) comprise images of tissues with NASH that have been treated with a drug (e.g., 50 uM ACC inhibitor or firsocostate). In some embodiments, the three groups of images capture tissues of the same subject at different times. In some embodiments, the three groups of images capture tissues of different subjects. The images can be phase images or fluorescence images generated using the techniques described herein (e.g., from bright-field images). Each phase or fluorescence image can be generated based on multiple bright-field images and thus comprises richer information such as phase shift information. Accordingly, the generated phase or fluorescence images can allow for higher precision in downstream analyses, as described below.

[0372] Specifically, to evaluate the efficacy of the drug, the system generates a distribution for each group of images to determine whether the distributions reflect an effect of the drug on the disease state. In FIG. 16C, three probability distributions are generated using a classifier trained on generated images. The X-axis indicates the probability outputted by the classifier upon receiving an input image (e.g., a generated phase image). As shown by the three distributions, generated images of healthy tissues (i.e., distribution 1620) are generally classified as having lower probabilities of having the disease than generated images of the diseased tissues (i.e., distribution 1624). Further, generated images of treated tissues (i.e., distribution 1622) are generally classified as having lower probabilities of having the disease than generated images of diseased tissues. A comparison of these distributions may indicate that the drug is effective at reversing or reducing the disease state, because the treatment has caused the diseased tissues to include features more similar to the healthy state and less similar to the diseased state. Thus, the synthetic images can enable biophysics understanding of the distribution of lipids in the cells and provide insight for treatment.

[0373] In some embodiments, rather than using distributions, the system can identify image clusters in an embedding space (e.g., UMAP), as shown in FIG. 16D. In the UMAP, each point represents an image embedding of an image (e.g., a generated phase image). As shown, the generated images of treated tissues form a cluster that moves away from the cluster of generated diseased images and toward the cluster of generated healthy images, which may indicate that the drug is effective at reversing or reducing the disease state. 2025287403   30 Dec 2025

[0374] The analysis in FIGS. 16C and 16D can applied to evaluate a plurality of drug candidates with respect to a disease of interest. For example, tissues treated by each drug candidate can be imaged, for example, by a microscope, to obtain bright-field images. The bright-field images can be transformed into phase images using the techniques described herein. Distribution or cluster can be generated for phase images of each treatment. Accordingly, the resulting plot can comprise a distribution or cluster representing the disease state, a distribution or cluster representing the health state, and a plurality of distributions or clusters each representing a candidate drug. The system can then identify the distribution or cluster closest to the healthy state to identify the most effective candidate drug candidate.

[0375] FIG. 17A illustrates synthetic images generated by an exemplary GAN model to study tuberous sclerosis (“TSC”), in accordance with some embodiments. In the depicted example, NGN2 neurons are illuminated using an optimal illumination pattern identified using techniques described herein, and bright-field images are captured. The bright-field images are inputted into a GAN model, which outputs generated phase images and generated fluorescence / bodipy images in FIG. 17A. The GAN model used in FIG. 17A can be configured to include a trained classifier as described with reference to FIGS. 14A-B and 15. The classifier has 2 conditions: a healthy condition (e.g., wild type) and a diseased condition (e.g., TSC KO). As described, the GAN model can be trained such that the generated images can be classified by the classifier to the same or similar degree of accuracy as real images.

[0376] The synthetic images generated by the GAN model can also be used to evaluate the efficacy of a treatment. The system processes three groups of generated images: a first group of images depicting healthy tissues that do not have the disease, a second group of images depicting untreated diseased tissues, and a third group of images depicting diseased tissues that have been treated (e.g., using a particular drug). In the depicted example in FIG. 17B, the first group of images (labelled “Wildtype”) comprises images of healthy tissues that do not have the TSC disease, the second group of images (labelled “TSC”) comprises images of untreated tissues having the TSC disease, and the third group of images (labelled “TSC+Rapamycin”) comprise images of tissues with TSC that have been treated with a drug. In some embodiments, the three groups of images capture tissues of the same subject at different times. In some embodiments, the three groups of images capture tissues of different subjects. The images are phase images generated using the techniques described herein (e.g., from bright-field images).

[0377] Specifically, to evaluate the efficacy of the drug, the system generates a distribution for each group of images to determine whether the distributions reflect an effect of the drug on the disease. In FIG. 17B, three biomarker distributions are generated. As shown by these 2025287403   30 Dec 2025 distributions, the drug appears to be effective at reversing or reducing the disease state, because the treatment has caused the diseased tissues to include features more similar to the healthy state and less similar to the diseased state. The analysis in FIG. 17B can applied to evaluate a plurality of drug candidates with respect to a disease of interest, as described above.

[0378] Exemplary methods, non-transitory computer-readable storage media, systems, and electronic devices are set out in the following items: 1. A method for training a machine-learning model to generate images of biological samples, comprising: obtaining a plurality of training images comprising: a training image of a first type, and a training image of a second type; generating, based on the training image of the first type, a plurality of wavelet coefficients using the machine-learning model; generating, based on the plurality of wavelet coefficients, a synthetic image of the second type; comparing the synthetic image of the second type with the training image of the second type; and updating the machine-learning model based on the comparison. 2. The method of item 1, wherein the training image of the first type is a bright-field image of a biological sample. 3. The method of item 2, wherein the training image of the second type is a fluorescence image of the biological sample. 4. The method of any of items 1-3, wherein the machine-learning model comprises a generator and a discriminator. 5. The method of item 4, wherein the machine-learning model comprises a conditional GAN model. 6. The method of any of items 4-5, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups. 7. The method of item 6, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group. 8. The method of any of items 6-7, wherein the plurality of neural networks comprises a plurality of U-Net neural networks. 9. The method of any of items 5-8, wherein the discriminator is a PatchGAN neural network. 2025287403   30 Dec 2025 10. The method of any of items 1-9, further comprising: generating, based on the training image of the first type, an image of a third type. 11. The method of item 10, wherein the image of the third type is a phase shift image. 12. The method of any of items 1-11, further comprising: generating, based on the training image of the first type, an image of a fourth type. 13. The method of item 12, wherein the image of the fourth type comprises segmentation data. 14. The method of any of items 1-13, wherein the training image of the first type is captured using a microscope according to a first illumination scheme. 15. The method of item 14, wherein the first illumination scheme comprises one or more illumination patterns. 16. The method of any of items 14-15, wherein the training image of the first type is part of a bright-field image array. 17. The method of any of items 14-16, wherein the plurality of training images is a first plurality of training images, the method further comprising: based on the comparison, identifying a second illumination scheme; obtaining a second plurality of training images comprising one or more images of the first type, wherein the one or more images of the first type are obtained based on the second illumination scheme; training the machine-learning model based on the second plurality of training images. 18. The method of any of items 1-16, further comprising: obtaining, using a microscope, a plurality of images of the first type; and generating, based on the obtained plurality of images, a plurality of synthetic images of the second type using the machine-learning model. 19. The method of item 18, further comprising: training a classifier based on the plurality of synthetic images of the second type. 20. The method of item 19, wherein the microscope is a first microscope, wherein the classifier is a first classifier, further comprising: obtaining, using a second microscope, a plurality of images of the second type; training a second classifier based on the plurality of images of the second type; comparing performance of the first classifier and the second classifier. 21. The method of item 20, wherein the second microscope is a fluorescence microscope. 2025287403   30 Dec 2025 22. A method for generating enhanced images of biological samples, comprising: obtaining, using a microscope, an image of a biological sample; and generating, based on the image, an enhanced image of the biological sample using a machine-learning model, wherein the machine-learning model has been trained by: obtaining a plurality of training images comprising: a training image of a first type, and a training image of a second type; generating, based on the training image of the first type, a plurality of wavelet coefficients using the machine-learning model; generating, based on the plurality of wavelet coefficients, a synthetic image of the second type; comparing the synthetic image of the second type with the training image of the second type; and updating the machine-learning model based on the comparison. 23. The method of item 22, wherein the training image of the first type is a bright-field image of a biological sample. 24. The method of item 22, wherein the training image of the second type is a fluorescence image of the biological sample. 25. The method of any of items 22-24, wherein the machine-learning model comprises a generator and a discriminator. 26. The method of item 25, wherein the machine-learning model comprises a conditional GAN model. 27. The method of any of items 25-26, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups. 28. The method of item 27, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group. 29. The method of any of items 27-28, wherein the plurality of neural networks comprises a plurality of U-Net neural networks. 30. The method of any of items 26-29, wherein the discriminator is a PatchGAN neural network. 31. The method of any of items 23-30, further comprising: generating, based on the training image of the first type, an image of a third type. 32. The method of item 31, wherein the image of the third type is a phase shift image. 2025287403   30 Dec 2025 33. The method of any of items 23-32, further comprising: generating, based on the training image of the first type, an image of a fourth type. 34. The method of item 33, wherein the image of the fourth type comprises segmentation data. 35. The method of any of items 23-34, wherein the training image of the first type is captured using a microscope according to a first illumination scheme. 36. The method of item 35, wherein the first illumination scheme comprises one or more illumination patterns. 37. The method of any of items 35-36, wherein the training image of the first type is part of a bright-field image array. 38. The method of any of items 35-37, wherein the plurality of training images is a first plurality of training images, the method further comprising: based on the comparison, identifying a second illumination scheme; obtaining a second plurality of training images comprising one or more images of the first type, wherein the one or more images of the first type are obtained based on the second illumination scheme; training the machine-learning model based on the second plurality of training images. 39. The method of any of items 35-38, further comprising: obtaining, using a microscope, a plurality of images of the first type; and generating, based on the obtained plurality of images, a plurality of synthetic images of the second type using the machine-learning model. 40. The method of item 39, further comprising: training a classifier based on the plurality of synthetic images of the second type. 41. The method of item 40, wherein the microscope is a first microscope, wherein the classifier is a first classifier, further comprising: obtaining, using a second microscope, a plurality of images of the second type; training a second classifier based on the plurality of images of the second type; comparing performance of the first classifier and the second classifier. 42. The method of item 41, wherein the second microscope is a fluorescence microscope. 43. A system for training a machine-learning model to generate images of biological samples, comprising: a computing system comprising one or more processors, and one or more memories storing a machine-learning model, wherein the computing system is configured to receive a 2025287403   30 Dec 2025 plurality of training images of a first type and one a training image of a second type, and wherein the computing system is configured to: generate, based on the training images of the first type, a plurality of wavelet coefficients using the machine-learning model; generate, based on the plurality of wavelet coefficients, a synthetic image of the second type; compare the synthetic image of the second type with the training image of the second type; and update the machine-learning model based on the comparison. 44. The system of item 43, wherein the training image of the first type is a bright-field image of a biological sample. 45. The system of any of items 43-44, wherein the training image of the second type is a fluorescence image of the biological sample. 46. The system of any of items 43-45, wherein the machine-learning model comprises a generator and a discriminator. 47. The system of item 46, wherein the machine-learning model comprises a conditional GAN model. 48. The system of any of items 46-47, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups. 49. The system of item 48, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group. 50. The system of any of items 48-49, wherein the plurality of neural networks comprises a plurality of U-Net neural networks. 51. The system of any of items 46-50, wherein the discriminator is a PatchGAN neural network. 52. The system of any of items 43-51, wherein the computing system is further configured to: generate, based on the training image of the first type, an image of a third type. 53. The system of item 52, wherein the image of the third type is a phase shift image. 54. The system of any of items 43-53, wherein the computing system is further configured to: generate, based on the training image of the first type, an image of a fourth type. 55. The system of item 54, wherein the image of the fourth type comprises segmentation data. 56. The system of any of items 43-55, wherein the training image of the first type is captured using a microscope according to a first illumination scheme. 2025287403   30 Dec 2025 57. The system of item 56, wherein the first illumination scheme comprises one or more illumination patterns. 58. The system of any of items 56-57, wherein the training image of the first type is part of a bright-field image array. 59. The system of any of items 56-58, wherein the plurality of training images is a first plurality of training images, and wherein the computing system is further configured to: based on the comparison, identify a second illumination scheme; obtain a second plurality of training images comprising one or more images of the first type, wherein the one or more images of the first type are obtained based on the second illumination scheme; train the machine-learning model based on the second plurality of training images. 60. The system of any of items 43-59, wherein the computing system is further configured to: obtain, using a microscope, a plurality of images of the first type; and generate, based on the obtained plurality of images, a plurality of synthetic images of the second type using the machine-learning model. 61. The system of item 59, wherein the computing system is further configured to: train a classifier based on the plurality of synthetic images of the second type. 62. The system of item 61, wherein the microscope is a first microscope, wherein the classifier is a first classifier, wherein the computing system is further configured to: obtain, using a second microscope, a plurality of images of the second type; train a second classifier based on the plurality of images of the second type; compare performance of the first classifier and the second classifier. 63. The system of item 62, wherein the second microscope is a fluorescence microscope. 64. A system for generating enhanced images of biological samples, comprising: a computing system comprising one or more processors, and one or more memories storing a machine-learning model, wherein the computing system is configured to receive an image of a biological sample obtained from a microscope and generate, based on the image, an enhanced image of the biological sample using a machine-learning model, wherein the machine-learning model has been trained by: obtaining a plurality of training images comprising: a training image of a first type, and a training image of a second type; 2025287403   30 Dec 2025 generating, based on the training image of the first type, a plurality of wavelet coefficients using the machine-learning model; generating, based on the plurality of wavelet coefficients, a synthetic image of the second type; comparing the synthetic image of the second type with the training image of the second type; and updating the machine-learning model based on the comparison. 65. The system of item 64, wherein the training image of the first type is a bright-field image of a biological sample. 66. The system of item 64, wherein the training image of the second type is a fluorescence image of the biological sample. 67. The system of any of items 64-66, wherein the machine-learning model comprises a generator and a discriminator. 68. The system of item 67, wherein the machine-learning model comprises a conditional GAN model. 69. The system of any of items 67-68, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups. 70. The system of item 69, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group. 71. The system of any of items 69-70, wherein the plurality of neural networks comprises a plurality of U-Net neural networks. 72. The system of any of items 67-71, wherein the discriminator is a PatchGAN neural network. 73.    The system of any of items 64-72, wherein the machine learning model is further trained by generating, based on the training image of the first type, an image of a third type. 74.    The system of item 73, wherein the image of the third type is a phase shift image. 75.    The system of any of items 64-74, wherein the machine-learning model has been trained by: generating, based on the training image of the first type, an image of a fourth type. 76. The system of item 75, wherein the image of the fourth type comprises segmentation data. 77. The system of any of items 64-76, wherein the training image of the first type is captured using a microscope according to a first illumination scheme. 78. The system of item 77, wherein the first illumination scheme comprises one or more illumination patterns. 2025287403   30 Dec 2025 79. The system of any of items 77-78, wherein the training image of the first type is part of a bright-field image array. 80. The system of any of items 77-79, wherein the plurality of training images is a first plurality of training images, wherein the machine-learning model has been trained by: based on the comparison, identifying a second illumination scheme; obtaining a second plurality of training images comprising one or more images of the first type, wherein the one or more images of the first type are obtained based on the second illumination scheme; training the machine-learning model based on the second plurality of training images. 81. The system of any of items 77-80, wherein the machine-learning model has been trained by: obtaining, using a microscope, a plurality of images of the first type; and generating, based on the obtained plurality of images, a plurality of synthetic images of the second type using the machine-learning model. 82. The system of item 81, wherein the machine-learning model has been trained by: training a classifier based on the plurality of synthetic images of the second type. 83. The system of item 82, wherein the microscope is a first microscope, wherein the classifier is a first classifier, wherein the machine-learning model has been trained by: obtaining, using a second microscope, a plurality of images of the second type; training a second classifier based on the plurality of images of the second type; comparing performance of the first classifier and the second classifier. 84. The system of item 83, wherein the second microscope is a fluorescence microscope. 85. A method of processing images of a biological sample to obtain one or more output images, comprising: obtaining a plurality of images of the biological sample using a plurality of configurations of a SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and inputting the plurality of images of the biological sample into a trained machinelearning model to obtain the one or more outputs images. 86. The method of item 85, wherein at least one configuration of the plurality of configurations of the SLM is to generate one or more optical aberrations. 87. The method of item 86, wherein generating one or more optical aberrations comprises a spherical aberration, astigmatism, defocus, distortion, tilt, or any combination thereof. 2025287403   30 Dec 2025 88. The method of any of items 85-86, wherein at least one configuration of the plurality of configurations of the SLM is to enhance one or more features. 89. The method of item 88, wherein the one or more features comprise a cell border, an actin filament, nuclear shape, cytoplasm segmentation, or any combination thereof. 90. The method of any of items 85-89, wherein at least one configuration of the plurality of configurations of the SLM is to reduce optical aberrations. 91. The method of any of items 85-90, wherein the plurality of SLM configurations is to obtain images of the biological sample at different depths. 92. The method of any of items 85-91, wherein the machine-learning model is configured to generate, based on an image of a first type, an image of a second type. 93.    The method of item 92, wherein the first type of images are bright-field images. 94.    The method of item 92, wherein the second type of images are fluorescence images. 95. The method of item 92, wherein the second type of images are enhanced versions of the first type of images. 96. The method of any of items 92-95, wherein the machine-learning model is a GAN model or a self-supervised model. 97. The method of any of items 85-96, wherein the plurality of images are obtained using a plurality of configurations of a light source of the optical system. 98.    The method of item 97, wherein the light source is a LED array of the optical system. 99.    The method of any of items 85-98, wherein at least one configuration of the plurality of SLM configurations is obtained by: training the machine-learning model; evaluating the trained machine-learning model; and identifying the at least one configuration based on the evaluation. 100. The method of any of items 85-99, wherein the trained machine-learning model is configured to receive an input image and output an enhanced version of the input image. 101. The method of item 100, wherein the enhanced version of the input image comprises one or more enhanced cellular phenotypes. 102.   An electronic device for processing images of a biological sample to obtain one or more output images, comprising: one or more processors; a memory; and 2025287403   30 Dec 2025 one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for: obtaining a plurality of images of the biological sample using a plurality of configurations of a SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and inputting the plurality of images of the biological sample into a trained machinelearning model to obtain the one or more output images. 103. A non-transitory computer-readable storage medium storing one or more programs for processing images of a biological sample to obtain one or more output images, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to: obtain a plurality of images of the biological sample using a plurality of configurations of a SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and input the plurality of images of the biological sample into a trained machine-learning model to obtain the one or more output images. 104. A method of classifying images of a biological sample, comprising: obtaining a plurality of images of the biological sample using a plurality of configurations of an SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and inputting the plurality of images of the biological sample into a trained machinelearning model to obtain one or more classification outputs. 105. The method of item 104, wherein at least one configuration of the plurality of configurations of the SLM is to generate one or more optical aberrations. 106. The method of item 105, wherein generating one or more optical aberrations comprises a spherical aberration, astigmatism, defocus, distortion, tilt, or any combination thereof. 107. The method of any of items 104-106, wherein at least one configuration of the plurality of configurations of the SLM is to enhance one or more features. 108. The method of item 107, wherein the one or more features comprise a cell border, an actin filament, nuclear shape, cytoplasm segmentation, or any combination thereof. 109. The method of any of items 104-108, wherein at least one configuration of the plurality of configurations of the SLM is to reduce optical aberrations. 2025287403   30 Dec 2025 110. The method of any of items 104-109, wherein the plurality of SLM configurations is to obtain images of the biological sample at different depths. 111. The method of any of items 104-110, wherein the plurality of images are obtained using a plurality of configurations of a light source of the optical system. 112. The method of item 111, wherein the light source is a LED array of the optical system. 113. The method of any of items 104-112, wherein at least one configuration of the plurality of SLM configurations is obtained by: training the machine-learning model; evaluating the trained machine-learning model; and identifying the at least one configuration based on the evaluation. 114. The method of any of items 104-113, wherein the trained machine-learning model is configured to receive an input image and detect one or more pre-defined objects in the input image. 115. The method of item 114, wherein the pre-defined objects include a diseased tissue. 116. An electronic device for classifying images of a biological sample, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for: obtaining a plurality of images of the biological sample using a plurality of configurations of an SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and inputting the plurality of images of the biological sample into a trained machinelearning model to obtain one or more classification outputs. 117. A non-transitory computer-readable storage medium storing one or more programs for classifying images of a biological sample, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to: obtain a plurality of images of the biological sample using a plurality of configurations of an SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and input the plurality of images of the biological sample into a trained machine-learning model to obtain one or more classification outputs. 2025287403   30 Dec 2025 118. A method for training a machine-learning model, comprising: obtaining a plurality of images of a biological sample using a plurality of configurations of an SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and training the machine-learning model using the plurality of images. 119. The method of item 118, wherein at least one configuration of the plurality of configurations of the SLM is to generate one or more optical aberrations. 120. The method of item 119, wherein generating one or more optical aberrations comprises a spherical aberration, astigmatism, defocus, distortion, tilt, or any combination thereof. 121. The method of any of items 118-120, wherein at least one configuration of the plurality of configurations of the SLM is to enhance one or more features. 122. The method of item 121, wherein the one or more features comprise a cell border, an actin filament, nuclear shape, cytoplasm segmentation, or any combination thereof. 123. The method of any of items 118-122, wherein at least one configuration of the plurality of configurations of the SLM is to reduce optical aberrations. 124. The method of any of items 118-123, wherein at least one configuration of the plurality of configurations of the SLM is to obtain images of the biological sample at different depths. 125. The method of any of items 118-124, wherein the machine-learning model is configured to generate, based on an image of a first type, an image of a second type. 126. The method of item 125, wherein the first type of images are bright-field images. 127. The method of item 125, wherein the second type of images are fluorescence images. 128. The method of any of items 118-127, wherein the machine-learning model is a GAN model or a self-supervised model. 129. The method of any of items 118-128, wherein the machine-learning model is a classification model. 130. The method of any of items 118-129, wherein the plurality of images are obtained using a plurality of configurations of a light source of the optical system. 131. The method of item 130, wherein the light source is a LED array of the optical system. 132. The method of any of items 118-131, wherein training the machine-learning model comprises: (a) training the machine-learning model using a first image, wherein the first image is obtained using a first configuration of the SLM of the optical system; (b) evaluating the trained machine-learning model; (c) based on the evaluation, identifying a second configuration of the SLM; and 2025287403   30 Dec 2025 (d) training the machine-learning model using a second image, wherein the second image is obtained using the second configuration of the SLM of the optical system. 133. The method of item 112, wherein the evaluation is based on a loss function of the machine-learning model. 134. The method of any of items 112-113, further comprising: repeating steps (a)-(d) until a threshold is met. 135. The method of item 114, wherein the threshold is indicative of convergence of the training. 136. The method of any of items 118-135, wherein the trained machine-learning model is configured to receive an input image and output an enhanced version of the input image. 137. The method of item 136, wherein the enhanced version of the input image comprises one or more enhanced cellular phenotypes. 138. The method of any of items 118-135, wherein the trained machine-learning model is configured to receive an input image and detect one or more pre-defined objects in the input image. 139. The method of item 138, wherein the pre-defined objects include a diseased tissue. 140. An electronic device for training a machine-learning model, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for: obtaining a plurality of images of a biological sample using a plurality of configurations of an SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and training the machine-learning model using the plurality of images. 141. A non-transitory computer-readable storage medium storing one or more programs for training a machine-learning model, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to: obtain a plurality of images of a biological sample using a plurality of configurations of an SLM of an optical system, wherein the SLM is located in an optical path between the biological sample and an image recording device; and train the machine-learning model using the plurality of images. 2025287403   30 Dec 2025 142. A method of generating enhanced images of biological samples, comprising: obtaining, using a microscope, an image of a biological sample illuminated using an illumination pattern of an illumination source, wherein the illumination pattern is determined by: training a classification model configured to receive an input image and output a classification result, training, using the trained classification model, a machine-learning model having an plurality of weights corresponding to a plurality of illumination settings, and identifying the illumination pattern based on the plurality of weights of the trained machine-learning model; and generating an enhanced image of the biological sample by inputting the obtained image of the biological sample into the trained machine-learning model. 143. The method of item 142, wherein the obtained image is a bright-field image. 144. The method of any of items 142-143, wherein the enhanced image is a fluorescence image, a phase image, or a combination thereof. 145. The method of any of items 142-144, wherein the illumination source comprises an array of illumination emitters. 146. The method of item 145, wherein the illumination source is a LED array. 147. The method of any of items 142-146, wherein the illumination pattern indicates whether each illumination emitter is turned on or off and the intensity of each illumination emitter. 148. The method of any of items 145-147, wherein each illumination setting of the plurality of illumination settings corresponds to a respective illumination emitter of the illumination source; and wherein each weight corresponds to an intensity of the respective illumination emitter. 149. The method of any of items 142-148, wherein the classification model is configured to receive an input phase image or an input fluorescence image and output a classification result indicative of one class out of a plurality of pre-defined classes. 150. The method of item 149, wherein the plurality of pre-defined classes comprises a healthy class and a diseased class. 151. The method of item 150, wherein the machine-learning model is a GAN model comprising an attention layer comprising the plurality of weights, a discriminator, and a generator. 2025287403   30 Dec 2025 152. The method of item 151, wherein the machine-learning model is a conditional GAN model. 153. The method of any of items 151-152, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups. 154. The method of item 153, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group. 155. The method of any of items 153-154, wherein the plurality of neural networks comprises a plurality of U-Net neural networks. 156. The method of any of items 151-155, wherein the discriminator is a PatchGAN neural network. 157. The method of any of items 151-156, wherein training, using the trained classification model, the machine-learning model comprises: applying the plurality of weights to a plurality of bright-field training images; aggregating the plurality of weighted bright-field training images into an aggregated bright-field image; inputting the aggregated bright-field training image into the machine-learning model to obtain an enhanced training image and a generator loss; inputting the enhanced training image into the trained classifier to obtain a classifier loss; augmenting the generator loss based on the classifier loss; and updating the plurality of weights based on the augmented generator loss. 158. The method of any of items 142-157, further comprising: classifying the enhanced image using the trained classifier. 159. The method of any of items 142-158, further comprising: displaying the enhanced image. 160. A system for generating enhanced images of biological samples, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for: obtaining, using a microscope, an image of a biological sample illuminated using an illumination pattern of an illumination source, wherein the illumination pattern is determined by: 2025287403   30 Dec 2025 training a classification model configured to receive an input image and output a classification result, training, using the trained classification model, a machine-learning model having an plurality of weights corresponding to a plurality of illumination settings, and identifying the illumination pattern based on the plurality of weights of the trained machine-learning model; and generating an enhanced image of the biological sample by inputting the obtained image of the biological sample into the trained machine-learning model. 161. A non-transitory computer-readable storage medium storing one or more programs for generating enhanced images of biological samples, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to: obtain, using a microscope, an image of a biological sample illuminated using an illumination pattern of an illumination source, wherein the illumination pattern is determined by: training a classification model configured to receive an input image and output a classification result, training, using the trained classification model, a machine-learning model having an plurality of weights corresponding to a plurality of illumination settings, and identifying the illumination pattern based on the plurality of weights of the trained machine-learning model; and generate an enhanced image of the biological sample by inputting the obtained image of the biological sample into the trained machine-learning model. 162. A method of evaluating a treatment with respect to a disease of interest, comprising: receiving a first plurality of images depicting a first set of healthy biological samples not affected by the disease of interest; receiving a second plurality of images depicting a second set of untreated biological samples affected by the disease of interest; receiving a third plurality of images depicting a third set of treated biological samples affected by the disease of interest and treated by the treatment; inputting the first plurality of images into a trained machine-learning model to obtain a first plurality of enhanced images; inputting the second plurality of images into the trained machine-learning model to obtain a second plurality of enhanced images; 2025287403   30 Dec 2025 inputting the third plurality of images into the trained machine-learning model to obtain a third plurality of enhanced images; comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment. 163. The method of item 162, wherein the first plurality of images, the second plurality of images, and the third plurality of images are bright-field images. 164. The method of any of items 162-163, wherein the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images are fluorescence images. 165. The method of any of items 162-163, wherein the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images are phase images. 166. The method of any of items 162-165, wherein comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment comprises: identifying, in each image, a signal associated with a biomarker. 167. The method of item 166, wherein comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment further comprises: determining a first distribution based on signals of the biomarker in the first plurality of enhanced images; determining a second distribution based on signals of the biomarker in the second plurality of enhanced images; and determining a third distribution based on signals of the biomarker in the third plurality of enhanced images. 168. The method of item 167, wherein comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment further comprises: comparing the first distribution, the second distribution, and the third distribution to evaluate the treatment. 169. The method of any of items 162-165, wherein comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment comprises: determining, for each image, a score indicative of the statement of the disease of interest. 2025287403   30 Dec 2025 170. The method of item 169, wherein comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment further comprises: determining a first distribution based on scores of the first plurality of enhanced images; determining a second distribution based on scores of the second plurality of enhanced images; and determining a third distribution based on scores of the third plurality of enhanced images. 171. The method of item 170, wherein comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment further comprises: comparing the first distribution, the second distribution, and the third distribution to evaluate the treatment. 172. The method of any of items 162-171, wherein the treatment is a first treatment, the method further comprising: receiving a fourth plurality of images depicting a fourth set of treated biological samples affected by the disease of interest and treated by a second treatment; inputting the fourth plurality of images into the trained machine-learning model to obtain a fourth plurality of enhanced images; comparing the first plurality of enhanced images, the second plurality of enhanced images, the third plurality of enhanced images, and the fourth plurality of enhanced images to compare the first treatment and the second treatment. 173. The method of item 172, further comprising: selecting a treatment out of the first treatment and the second treatment based on the comparison. 174. The method of item 173, further comprising: administering the selected treatment. 175. The method of item 173, further comprising: providing a medical recommendation based on the selected treatment. 176. The method of any of items 162-175, wherein the trained machine-learning model is is a GAN model comprising a discriminator and a generator. 177. The method of item 176, wherein the machine-learning model is a conditional GAN model. 178. The method of any of items 176-177, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups. 2025287403   30 Dec 2025 179. The method of item 178, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group. 180. The method of any of items 176-179, wherein the discriminator is a PatchGAN neural network. 181. A system for evaluating a treatment with respect to a disease of interest, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for: receiving a first plurality of images depicting a first set of healthy biological samples not affected by the disease of interest; receiving a second plurality of images depicting a second set of untreated biological samples affected by the disease of interest; receiving a third plurality of images depicting a third set of treated biological samples affected by the disease of interest and treated by the treatment; inputting the first plurality of images into a trained machine-learning model to obtain a first plurality of enhanced images; inputting the second plurality of images into the trained machine-learning model to obtain a second plurality of enhanced images; inputting the third plurality of images into the trained machine-learning model to obtain a third plurality of enhanced images; comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment. 182. A non-transitory computer-readable storage medium storing one or more programs for evaluating a treatment with respect to a disease of interest, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to: receiving a first plurality of images depicting a first set of healthy biological samples not affected by the disease of interest; receiving a second plurality of images depicting a second set of untreated biological samples affected by the disease of interest; receiving a third plurality of images depicting a third set of treated biological samples affected by the disease of interest and treated by the treatment; 2025287403   30 Dec 2025 inputting the first plurality of images into a trained machine-learning model to obtain a first plurality of enhanced images; inputting the second plurality of images into the trained machine-learning model to obtain a second plurality of enhanced images; inputting the third plurality of images into the trained machine-learning model to obtain a third plurality of enhanced images; comparing the first plurality of enhanced images, the second plurality of enhanced images, and the third plurality of enhanced images to evaluate the treatment.

[0379] Although the disclosure and examples have been fully described with reference to the accompanying figures, it is to be noted that various changes and modifications will become apparent to those skilled in the art. Such changes and modifications are to be understood as being included within the scope of the disclosure and examples as defined by the claims.

[0380] The foregoing description, for purpose of explanation, has been described with reference to specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. The embodiments were chosen and described in order to best explain the principles of the techniques and their practical applications. Others skilled in the art are thereby enabled to best utilize the techniques and various embodiments with various modifications as are suited to the particular use contemplated.

[0381] Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

[0382] The reference to any prior art in this specification is not, and should not be taken as, an acknowledgement or any form of suggestion that the prior art forms part of the common general knowledge in Australia.

Claims

1. A system for generating enhanced images of biological samples, the system comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for:providing as input to a machine-learning model at least one label-free microscopic image of a biological sample, wherein the at least one label-free microscopic image requires no exogenous biological or chemical labels or stains of the biological sample, and wherein the at least one label-free microscopic image comprises an image that is not a brightfield image; andgenerating, based on the at least one label-free microscopic image of the biological sample, a labeled image of the biological sample using the machine-learning model, wherein the labeled image of the biological sample shows exogenous biological or chemical labels or stains of the biological sample, and wherein the labeled image is selected from the group consisting of: fluorescence images, genetically encoded fluorescent protein images, in situ hybridization images, bioluminescence images, chemiluminescence images, images of fluorophore-conjugated small molecule or peptide probes, nanoparticle-labeled images, semantic map images, absorbance map images, or phospho-specific immunostaining images.

2. The system of claim 1, wherein the at least one label-free microscopic image comprises one or more of: phase images, phase-shift images, quantitative phase images, differential interference contrast images, darkfield images, polarized light images, reflection interference contrast images, Raman microscopy images, coherent anti-stokes Raman scattering images, stimulated Raman scattering microscopy images, second-harmonic generation and third-harmonic generation microscopy images, infrared / Fourier transform infrared microscopy images, photoacoustic images, electron microscopy images, atomic force microscopy images, semantic maps, polarization maps, refractive maps (2D and 3D), or absorbance maps.2025287403   30 Dec 20253. The system of claim 1, wherein the fluorescence images comprise one or more of: DAPI images, MAP2 images, pS6 images, or immunofluorescence images.

4. The system of claim 1, wherein the fluorescence microscopy images comprise confocal, total internal reflection fluorescence, two-photon, and super-resolution images.

5. The system of claim 1, wherein the phospho-specific immunostaining images comprise one or more of: immunofluorescence images with phospho-specific antibodies, western blot images with fluorescent or chemiluminescent detection, images with genetically encoded phosphorylation reporters including FRET or BRET sensors, images with small-molecule fluorescent or bioluminescent probes, proximity ligation images, and immuno-gold labeled electron microscopy images.

6. The system of claim 1, wherein the at least one label-free microscopic image comprises images of two or more modalities.

7. The system of claim 6, wherein the machine learning model comprises two or more generators.

8. The system of claim 7, wherein the two or more generators generate two or more labeled, synthetic images from the label-free images of two or more modalities.

9. The system of claim 6, wherein the at least one label-free microscopic image comprises a brightfield image.

10. The system of claim 1, wherein the machine-learning model has been trained by:obtaining a training image of a first type and a training image of a second type, wherein the training image of the first type comprises a phase image;2025287403   30 Dec 2025generating, based on the training image of the first type, a synthetic image of the second type;comparing the synthetic image of the second type with the training image of the second type; andupdating the machine-learning model based on the comparison.

11. The system of claim 10, wherein:the training image of the first type comprises two or more images of different image modalities;the machine learning model comprises two or more generators; andthe two or more generators generate two or more synthetic images of the second type from the two or more images of different image modalities of the first type.

12. The system of claim 1, wherein the at least one label-free microscopic image comprises a phase shift image.

13. The system of claim 12, wherein each pixel of the phase shift image indicates a local value of a phase in the image.

14. The system of claim 1, wherein the machine-learning model comprises a generator and a discriminator.

15. The system of claim 14, wherein the machine-learning model comprises a conditional GAN model.

16. The system of claim 14, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups.2025287403   30 Dec 202517. The system of claim 16, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group.

18. The system of claim 16, wherein the plurality of neural networks comprises a plurality of U-Net neural networks.

19. The system of claim 15, wherein the discriminator is a PatchGAN neural network.

20. The system of claim 10, wherein the training image of the first type comprises animage captured using a microscope according to a first illumination scheme.

21. The system of claim 20, wherein the first illumination scheme comprises one or more illumination patterns.

22. The system of claim 21, wherein the one or more programs include instructions for:based on the comparison, identifying a second illumination scheme;obtaining at least one additional training image of the first type, wherein the at least one additional training image of the first type comprises at least one image obtained based on the second illumination scheme; andtraining the machine-learning model based on the at least one additional training image of the first type.

23. The system of claim 21, wherein the one or more programs include instructions for:obtaining a plurality of images of the first type; andgenerating, based on the obtained plurality of images, a plurality of synthetic images of the second type using the machine-learning model.2025287403   30 Dec 202524. The system of claim 23, wherein the one or more programs include instructions for: training a classifier based on the plurality of synthetic images of the second type.

25. A method for generating enhanced images of biological samples, the method comprising:providing as input to a machine-learning model at least one label-free microscopic image of a biological sample, wherein the at least one label-free microscopic image requires no exogenous biological or chemical labels or stains of the biological sample, and wherein the at least one label-free microscopic image comprises an image that is not a brightfield image; andgenerating, based on the at least one label-free microscopic image of the biological sample, a labeled image of the biological sample using the machine-learning model, wherein the labeled image of the biological sample shows exogenous biological or chemical labels or stains of the biological sample, and wherein the labeled image is selected from the group consisting of: fluorescence images, genetically encoded fluorescent protein images, in situ hybridization images, bioluminescence images, chemiluminescence images, images of fluorophore-conjugated small molecule or peptide probes, nanoparticle-labeled images, semantic map images, absorbance map images, or phospho-specific immunostaining images.

26. A non-transitory computer-readable storage medium storing one or moreprograms for generating enhanced images of biological samples, the one or more programs comprising instructions, which when executed by one or more processors of a system, cause the system to:provide as input to a machine-learning model at least one label-free microscopic image of a biological sample, wherein the at least one label-free microscopic image requires no exogenous biological or chemical labels or stains of the biological sample, and wherein the at least one label-free microscopic image comprises an image that is not a brightfield image; and2025287403   30 Dec 2025generate, based on the at least one label-free microscopic image of the biological sample, a labeled image of the biological sample using the machine-learning model, wherein the labeled image of the biological sample shows exogenous biological or chemical labels or stains of the biological sample, and wherein the labeled image is selected from the group consisting of: fluorescence images, genetically encoded fluorescent protein images, in situ hybridization images, bioluminescence images, chemiluminescence images, images of fluorophore-conjugated small molecule or peptide probes, nanoparticle-labeled images, semantic map images, absorbance map images, or phospho-specific immunostaining images.

27. The system of claim 1, wherein the in situ hybridization images comprise fluorescence in situ hybridization images.

28. A system for generating enhanced images of biological samples, the system comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for:providing as input to a trained machine-learning model at least two microscopic images of a biological sample, wherein the at least two microscopic images of the biological sample comprises a phase image, wherein the at least two microscopic images of the biological sample further comprise a bright-field image, a fluorescence image, a semantic map, a polarization map, a refractive map, an absorbance map, or any combination thereof; andgenerating, based on the at least two microscopic images of the biological sample, an enhanced image of the biological sample using the trained machine-learning model.

29. The system of claim 28, wherein the trained machine-learning model has been trained by:obtaining a training image of a first type and a training image of a second type, wherein the training image of the first type comprises a phase image;2025287403   30 Dec 2025generating, based on the training image of the first type, a synthetic image of the second type;comparing the synthetic image of the second type with the training image of the second type; andupdating the trained machine-learning model based on the comparison.

30. The system of claim 28, wherein the enhanced image of the biological sample comprises at least one of a bright-field image, a fluorescence image, a semantic map, a polarization map, a refractive map, and an absorbance map.

31. The system of claim 30, wherein the enhanced image of the biological sample comprises a fluorescence image.

32. The system of claim 28, wherein the phase image comprises a phase shift image.

33. The system of claim 32, wherein each pixel of the phase shift image indicates a local value of a phase in the image.

8. The system of claim 28, wherein the trained machine-learning model comprises a generator and a discriminator.

35. The system of claim 34, wherein the trained machine-learning model comprises a conditional GAN model.

36. The system of claim 34, wherein the generator comprises a plurality of neural networks corresponding to a plurality of frequency groups.2025287403   30 Dec 202537. The system of claim 36, wherein each neural network of the plurality of neural networks is configured to generate wavelet coefficients for a respective frequency group.

38. The system of claim 36, wherein the plurality of neural networks comprises a plurality of U-Net neural networks.

39. The system of claim 35, wherein the discriminator is a PatchGAN neural network.

40. The system of claim 29, wherein the training image of the first type comprises an image captured using a microscope according to a first illumination scheme.

41. The system of claim 40, wherein the first illumination scheme comprises one or more illumination patterns.

42. The system of claim 40, wherein the one or more programs include instructions for:based on the comparison, identifying a second illumination scheme;obtaining at least one additional training image of the first type, wherein the at least one additional training image of the first type comprises at least one image obtained based on the second illumination scheme; andtraining the trained machine-learning model based on the at least one additional training image of the first type.

43. The system of claim 40, wherein the one or more programs include instructions for:obtaining a plurality of images of the first type; and2025287403   30 Dec 2025generating, based on the obtained plurality of images, a plurality of synthetic images of the second type using the trained machine-learning model.

44. The system of claim 43, wherein the one or more programs include instructions for: training a classifier based on the plurality of synthetic images of the second type.

45. A method for generating enhanced images of biological samples, the method comprising:providing as input to a trained machine-learning model at least two microscopic images of a biological sample, wherein the at least two microscopic images of the biological sample comprises a phase image, wherein the at least two microscopic images of the biological sample further comprise a bright-field image, a fluorescence image, a semantic map, a polarization map, a refractive map, an absorbance map, or any combination thereof; andgenerating, based on the at least two microscopic images of the biological sample, an enhanced image of the biological sample using the trained machine-learning model.

46. A non-transitory computer-readable storage medium storing one or more programs for generating enhanced images of biological samples, the one or more programs comprising instructions, which when executed by one or more processors of a system, cause the system to:provide as input to a trained machine-learning model at least two microscopic images of a biological sample, wherein the at least two microscopic images of the biological sample comprises a phase image, wherein the at least two microscopic images of the biological sample further comprise a bright-field image, a fluorescence image, a semantic map, a polarization map, a refractive map, an absorbance map, or any combination thereof; andgenerate, based on the at least two microscopic images of the biological sample, an enhanced image of the biological sample using the trained machine-learning model.2025287403   30 Dec 202547. The system of claim 28, wherein the instructions for providing input to the trained machine-learning model comprise instructions for providing the phase image and the bright-field image to the trained machine-learning model.

48. The system of claim 28, wherein the instructions for providing input to the trained machine-learning model comprise instructions for providing to the trained machine-learning model (a) the phase image and (b) the bright-field image, the fluorescence image, the semantic map, the refractive map, or the absorbance map.

Images