arXiv subjects

Dushan N. Wadduwage

Publications and source records attributed to Dushan N. Wadduwage.

15 recordsLinked to original sources

arXiv2026

Beyond the Thin-Layer Limit: Differentiable Volumetric Training for Visible-Range Diffractive Neural Networks

Diffractive deep neural networks (D2NNs) promise miniaturized, power-efficient, light-speed optical front-ends for machine vision, yet the most mature demonstrations remain in the terahertz regime, built from readily fabricated millimeter-scale neurons. Translating D2NNs to the visible range, where nearly all vision pipelines operate, was long blamed on the difficulty of fabricating nanoscale neurons; but even after recent advances removed that barrier, visible-range D2NNs matching their terahertz counterparts remain out of reach. We identify the true obstacle as the thin-layer approximation underlying nearly all D2NN training, which treats each diffractive layer as an infinitely thin mask. It fails not because of the short wavelength, as is commonly assumed, but because the low-refractive-index materials (n approximately 1.3-1.5) used at visible wavelengths require relief structures thick enough that intra-layer diffraction and phase accumulation become significant. To overcome this, we introduce a differentiable beam-propagation ($\partial$BPM) layer that models each element as a finite-thickness volume and propagates light through it during training, keeping the fabrication-compatible height map end-to-end trainable without full-wave simulation in the loop. Across MNIST, Fashion-MNIST, and CIFAR-100 classification and imaging, $\partial$BPM training substantially reduces the design-to-device mismatch, and full-wave FDTD validation raises classification accuracy from 50% to 90% without re-optimization. The $\partial$BPM layer thus offers a scalable, physics-aware bridge between efficient optical neural-network optimization and fabrication-consistent diffractive design.

physics.optics↗

arXiv2026

Motion-guided sparse correction enables expert-quality point tracking across diverse microscopy regimes

Tracking the dynamics of non-canonical biological systems in microscopy videos remains a persistent challenge. Both classical and learning-based trackers depend on expert-reviewed data to be evaluated and adapted, yet exhaustive manual annotation rarely scales to the videos where these tools are needed most. We developed RIPPLE (Refinement Interpolation Platform for Point Location Estimation), which recasts annotation as sparse correction: a user clicks a starting point, RIPPLE proposes a full trajectory, and the user intervenes only where the trajectory drifts. We tested RIPPLE on five challenging microscopy datasets from our laboratories, four from the transparent jellyfish Clytia hemisphaerica and one tracking landmarks on rapidly moving sperm. Across these, RIPPLE matched the quality of exhaustive manual annotation while reducing manual clicks by 3 to 25 times across datasets. RIPPLE thereby fills a missing layer between manual annotation and fully automated tracking, enabling immediate quantification of biological dynamics, method benchmarking, and the production of the gold-standard data needed to adapt future automated microscopy trackers.

cs.CV↗

arXiv2026

Differentiable Microscopy Designs an All Optical Phase Retrieval Microscope

Designing new optical systems from the ground up for microscopy imaging tasks such as phase retrieval, requires substantial scientific expertise and creativity. To augment the traditional design process, we propose differentiable microscopy ($\partialμ$), which introduces a top-down design approach. Using all optical phase retrieval as an illustrative example, we demonstrate the effectiveness of data-driven microscopy design through $\partialμ$. Furthermore, we conduct comprehensive comparisons with existing all-optical phase retrieval methods, showcasing the consistent superiority of our learned designs across multiple datasets, including biological samples. To substantiate our ideas, we experimentally validate the functionality of one of the learned designs, providing a proof of concept. The proposed differentiable microscopy framework supplements the creative process of designing new phase microscopy systems and may be extended to other similar applications in optical design.

physics.optics↗

arXiv2026

Beyond Perfect Scores: Proof-by-Contradiction for Trustworthy Machine Learning

Machine learning (ML) models show strong promise for new biomedical prediction tasks, but concerns about trustworthiness have hindered their clinical adoption. In particular, it is often unclear whether a model relies on true clinical cues or on spurious hierarchical correlations in the data. This paper introduces a simple yet broadly applicable trustworthiness test grounded in stochastic proof-by-contradiction. Instead of just showing high test performance, our approach trains and tests on spurious labels carefully permuted based on a potential outcomes framework. A truly trustworthy model should fail under such label permutation; comparable accuracy across real and permuted labels indicates overfitting, shortcut learning, or data leakage. Our approach quantifies this behavior through interpretable Fisher-style p-values, which are well understood by domain experts across medical and life sciences. We evaluate our approach on multiple new bacterial diagnostics to separate tasks and models learning genuine causal relationships from those driven by dataset artifacts or statistical coincidences. Our work establishes a foundation to build rigor and trust between ML and life-science research communities, moving ML models one step closer to clinical adoption.

cs.LG↗

arXiv2025

Thinking in Groups: Permutation Tests Reveal Near-Out-of-Distribution

Deep neural networks (DNNs) have the potential to power many biomedical workflows, but training them on truly representative, IID datasets is often infeasible. Most models instead rely on biased or incomplete data, making them prone to out-of-distribution (OoD) inputs that closely resemble in-distribution samples. Such near-OoD cases are harder to detect than standard OOD benchmarks and can cause unreliable, even catastrophic, predictions. Biomedical assays, however, offer a unique opportunity: they often generate multiple correlated measurements per specimen through biological or technical replicates. Exploiting this insight, we introduce Homogeneous OoD (HOoD), a novel OoD detection framework for correlated data. HOoD projects groups of correlated measurements through a trained model and uses permutation-based hypothesis tests to compare them with known subpopulations. Each test yields an interpretable p-value, quantifying how well a group matches a subpopulation. By aggregating these p-values, HOoD reliably identifies OoD groups. In evaluations, HOoD consistently outperforms point-wise and ensemble-based OoD detectors, demonstrating its promise for robust real-world deployment.

cs.LG↗

arXiv2025

Uncertainty Awareness Enables Efficient Labeling for Cancer Subtyping in Digital Pathology

Machine-learning-assisted cancer subtyping is a promising avenue in digital pathology. Cancer subtyping models, however, require careful training using expert annotations so that they can be inferred with a degree of known certainty (or uncertainty). To this end, we introduce the concept of uncertainty awareness into a self-supervised contrastive learning model. This is achieved by computing an evidence vector at every epoch, which assesses the model's confidence in its predictions. The derived uncertainty score is then utilized as a metric to selectively label the most crucial images that require further annotation, thus iteratively refining the training process. With just 1-10% of strategically selected annotations, we attain state-of-the-art performance in cancer subtyping on benchmark datasets. Our method not only strategically guides the annotation process to minimize the need for extensive labeled datasets, but also improves the precision and efficiency of classifications. This development is particularly beneficial in settings where the availability of labeled data is limited, offering a promising direction for future research and application in digital pathology.

cs.CV↗

arXiv2023

From Hours to Seconds: Towards 100x Faster Quantitative Phase Imaging via Differentiable Microscopy

With applications ranging from metabolomics to histopathology, quantitative phase microscopy (QPM) is a powerful label-free imaging modality. Despite significant advances in fast multiplexed imaging sensors and deep-learning-based inverse solvers, the throughput of QPM is currently limited by the speed of electronic hardware. Complementarily, to improve throughput further, here we propose to acquire images in a compressed form such that more information can be transferred beyond the existing electronic hardware bottleneck. To this end, we present a learnable optical compression-decompression framework that learns content-specific features. The proposed differentiable quantitative phase microscopy ($\partial μ$) first uses learnable optical feature extractors as image compressors. The intensity representation produced by these networks is then captured by the imaging sensor. Finally, a reconstruction network running on electronic hardware decompresses the QPM images. In numerical experiments, the proposed system achieves compression of $\times$ 64 while maintaining the SSIM of $\sim 0.90$ and PSNR of $\sim 30$ dB on cells. The results demonstrated by our experiments open up a new pathway for achieving end-to-end optimized (i.e., optics and electronic) compact QPM systems that may provide unprecedented throughput improvements.

eess.IV↗

arXiv2023

Contrastive Deep Encoding Enables Uncertainty-aware Machine-learning-assisted Histopathology

Deep neural network models can learn clinically relevant features from millions of histopathology images. However generating high-quality annotations to train such models for each hospital, each cancer type, and each diagnostic task is prohibitively laborious. On the other hand, terabytes of training data -- while lacking reliable annotations -- are readily available in the public domain in some cases. In this work, we explore how these large datasets can be consciously utilized to pre-train deep networks to encode informative representations. We then fine-tune our pre-trained models on a fraction of annotated training data to perform specific downstream tasks. We show that our approach can reach the state-of-the-art (SOTA) for patch-level classification with only 1-10% randomly selected annotations compared to other SOTA approaches. Moreover, we propose an uncertainty-aware loss function, to quantify the model confidence during inference. Quantified uncertainty helps experts select the best instances to label for further training. Our uncertainty-aware labeling reaches the SOTA with significantly fewer annotations compared to random labeling. Last, we demonstrate how our pre-trained encoders can surpass current SOTA for whole-slide image classification with weak supervision. Our work lays the foundation for data and task-agnostic pre-trained deep networks with quantified uncertainty.

cs.CV↗

arXiv2023

MOSAIC: Masked Optimisation with Selective Attention for Image Reconstruction

Compressive sensing (CS) reconstructs images from sub-Nyquist measurements by solving a sparsity-regularized inverse problem. Traditional CS solvers use iterative optimizers with hand crafted sparsifiers, while early data-driven methods directly learn an inverse mapping from the low-dimensional measurement space to the original image space. The latter outperforms the former, but is restrictive to a pre-defined measurement domain. More recent, deep unrolling methods combine traditional proximal gradient methods and data-driven approaches to iteratively refine an image approximation. To achieve higher accuracy, it has also been suggested to learn both the sampling matrix, and the choice of measurement vectors adaptively. Contrary to the current trend, in this work we hypothesize that a general inverse mapping from a random set of compressed measurements to the image domain exists for a given measurement basis, and can be learned. Such a model is single-shot, non-restrictive and does not parametrize the sampling process. To this end, we propose MOSAIC, a novel compressive sensing framework to reconstruct images given any random selection of measurements, sampled using a fixed basis. Motivated by the uneven distribution of information across measurements, MOSAIC incorporates an embedding technique to efficiently apply attention mechanisms on an encoded sequence of measurements, while dispensing the need to use unrolled deep networks. A range of experiments validate our proposed architecture as a promising alternative for existing CS reconstruction methods, by achieving the state-of-the-art for metrics of reconstruction accuracy on standard datasets.

cs.CV↗

arXiv2022

DEEP$^2$: Deep Learning Powered De-scattering with Excitation Patterning

Limited throughput is a key challenge in in-vivo deep-tissue imaging using nonlinear optical microscopy. Point scanning multiphoton microscopy, the current gold standard, is slow especially compared to the wide-field imaging modalities used for optically cleared or thin specimens. We recently introduced 'De-scattering with Excitation Patterning or DEEP', as a widefield alternative to point-scanning geometries. Using patterned multiphoton excitation, DEEP encodes spatial information inside tissue before scattering. However, to de-scatter at typical depths, hundreds of such patterned excitations are needed. In this work, we present DEEP$^2$, a deep learning based model, that can de-scatter images from just tens of patterned excitations instead of hundreds. Consequently, we improve DEEP's throughput by almost an order of magnitude. We demonstrate our method in multiple numerical and physical experiments including in-vivo cortical vasculature imaging up to four scattering lengths deep, in alive mice.

eess.IV↗

arXiv2022

Deep Optical Coding Design in Computational Imaging

Computational optical imaging (COI) systems leverage optical coding elements (CE) in their setups to encode a high-dimensional scene in a single or multiple snapshots and decode it by using computational algorithms. The performance of COI systems highly depends on the design of its main components: the CE pattern and the computational method used to perform a given task. Conventional approaches rely on random patterns or analytical designs to set the distribution of the CE. However, the available data and algorithm capabilities of deep neural networks (DNNs) have opened a new horizon in CE data-driven designs that jointly consider the optical encoder and computational decoder. Specifically, by modeling the COI measurements through a fully differentiable image formation model that considers the physics-based propagation of light and its interaction with the CEs, the parameters that define the CE and the computational decoder can be optimized in an end-to-end (E2E) manner. Moreover, by optimizing just CEs in the same framework, inference tasks can be performed from pure optics. This work surveys the recent advances on CE data-driven design and provides guidelines on how to parametrize different optical elements to include them in the E2E framework. Since the E2E framework can handle different inference applications by changing the loss function and the DNN, we present low-level tasks such as spectral imaging reconstruction or high-level tasks such as pose estimation with privacy preserving enhanced by using optimal task-based optical architectures. Finally, we illustrate classification and 3D object recognition applications performed at the speed of the light using all-optics DNN.

cs.CV↗

arXiv2019

3D Deep Learning Enables Fast Imaging of Spines through Scattering Media by Temporal Focusing Microscopy

Today the gold standard for in vivo imaging through scattering tissue is the point-scanning two-photon microscope (PSTPM). Especially in neuroscience, PSTPM is widely used for deep-tissue imaging in the brain. However, due to sequential scanning, PSTPM is slow. Temporal focusing microscopy (TFM), on the other hand, focuses femtosecond pulsed laser light temporally, while keeping wide-field illumination, and is consequently much faster. However, due to the use of a camera detector, TFM suffers from the scattering of emission photons. As a result, TFM produces images of poor spatial resolution and signal-to-noise ratio (SNR), burying fluorescent signals from small structures such as dendritic spines. In this work, we present a data-driven deep learning approach to improve resolution and SNR of TFM images. Using a 3D convolutional neural network (CNN) we build a map from TFM to PSTPM modalities, to enable fast TFM imaging while maintaining high-resolution through scattering media. We demonstrate this approach for in vivo imaging of dendritic spines on pyramidal neurons in the mouse visual cortex. We show that our trained network rapidly outputs high-resolution images that recover biologically relevant features previously buried in the scattered fluorescence in the TFM images. In vivo imaging that combines TFM and the proposed 3D convolution neural network is one to two orders of magnitude faster than PSTPM but retains the high resolution and SNR necessary to analyze small fluorescent structures. The proposed 3D convolution deep network could also be potentially beneficial for improving the performance of many speed-demanding deep-tissue imaging applications such as in vivo voltage imaging.

eess.IV↗

arXiv2019

Multiline Orthogonal Scanning Temporal Focusing (mosTF) Microscopy for Scattering Reduction in High-speed in vivo Brain Imaging

Temporal focusing two-photon microscopy enables high resolution imaging of fine structures in vivo over a large volume. A limitation of temporal focusing is that signal-to-background ratio and resolution degrade rapidly with increasing imaging depth. This degradation originates from the scattered emission photons are widely distributed resulting in a strong background. We have developed Multiline Orthogonal Scanning Temporal Focusing (mosTF) microscopy that overcomes this problem. mosTF captures a sequence of images at each scan location of the excitation line, followed by a reconstruction algorithm reassigns scattered photons back to the correct scan position. We demonstrate mosTF by acquiring mice neuronal images in vivo. Our results show remarkably improvements with mosTF for in vivo brain imaging while maintaining its speed advantage.

physics.optics↗

arXiv2019

De-scattering with Excitation Patterning (DEEP) Enables Rapid Wide-field Imaging Through Scattering Media

From multi-photon imaging penetrating millimeters deep through scattering biological tissue, to super-resolution imaging conquering the diffraction limit, optical imaging techniques have greatly advanced in recent years. Notwithstanding, a key unmet challenge in all these imaging techniques is to perform rapid wide-field imaging through a turbid medium. Strategies such as active wave-front correction and multi-photon excitation, both used for deep tissue imaging; or wide-field total-internal-refection illumination, used for super-resolution imaging; can generate arbitrary excitation patterns over a large field-of-view through or under turbid media. In these cases, throughput advantage gained by wide-field excitation is lost due to the use of point detection. To address this challenge, here we introduce a novel technique called De-scattering with Excitation Patterning, or 'DEEP', which uses patterned excitation followed by wide-field detection with computational imaging. We use two-photon temporal focusing (TFM) to demonstrate our approach at multiple scattering lengths deep in tissue. Our results suggest that millions of point-scanning measurements could be substituted with tens to hundreds of DEEP measurements with no compromise in image quality.

physics.optics↗

arXiv2018

Chip-based Resonance Raman Spectroscopy Using Tantalum Pentoxide Waveguides

Blood analysis is an important diagnostic tool, as it provides a wealth of information about the patient health. Raman spectroscopy is a promising tool in blood analysis, but widespread clinical application is limited by its low signal strength, as well as complex and costly instrumentation. The growing field of waveguide-based Raman spectroscopy tries to solve these challenges by working towards fully integrated Raman sensors with increased interaction areas. In this work, we demonstrate on-chip resonance Raman measurements of hemoglobin, a crucial component of blood, at 532 nm excitation using a tantalum pentoxide (Ta2O5) waveguide platform. We have also characterized the background signal from Ta2O5 waveguide material when excited at 532 nm. In addition, we demonstrate spontaneous Raman measurements of iso-propanol and methanol using the same platform. Our results suggest that Ta2O5 is a promising waveguide platform for resonance Raman spectroscopy at 532 nm, and in particular for blood analysis.

physics.app-ph↗