Standard deep learning (DL) models often suffer significant performance degradation on out-of-distribution (OOD) data, where test data differs from training data, a common challenge in medical imaging due to real-world variations. We propose a unified self-censorship framework as an alternative to the standard DL models for glaucoma classification using deep evidential uncertainty quantification. Our approach detects OOD samples at both the dataset and image levels. Dataset-level self-censorship enables users to accept or reject predictions for an entire new dataset based on model uncertainty, whereas image-level self-censorship refrains from making predictions on individual OOD images rather than risking incorrect classifications. We validated our approach across diverse datasets. Our dataset-level self-censorship method outperforms the standard DL model in OOD detection, achieving an average 11.93% higher area under the curve (AUC) across 14 OOD datasets. Similarly, our image-level self-censorship model improves glaucoma classification accuracy by an average of 17.22% across 4 external glaucoma datasets against baselines while censoring 28.25% more data. Our approach addresses the challenge of generalization in standard DL models for glaucoma classification across diverse datasets by selectively withholding predictions when the model is uncertain. This method reduces misclassification errors compared to state-of-the-art baselines, particularly for OOD cases. This study introduces a tunable framework that explores the trade-off between prediction accuracy and data retention in glaucoma prediction. By managing uncertainty in model outputs, the approach lays a foundation for future decision support tools aimed at improving the reliability of automated glaucoma diagnosis.

Office-based testing, enhanced by advances in imaging technology, is routinely used in eye care to non-invasively assess ocular structure and function. This type of imaging coupled with autonomous artificial intelligence holds immense opportunity to diagnose eye diseases quickly. Despite the wide availability and use of ocular imaging, there are several factors that hinder optimization of clinical practice and patient care. While some large institutions have developed end-to-end digital workflows that utilize electronic health records, enterprise imaging archives, and dedicated diagnostic viewers, this experience has not yet made its way to smaller and independent eye clinics. Fractured interoperability practices impact patient care in all healthcare domains, including eye care where there is a scarcity of care centers, making collaboration essential among providers, specialists, and primary care who might be treating systemic conditions with profound impact on vision. The purpose of this white paper is to describe the current state of ocular imaging by focusing on the challenges related to interoperability, reporting, and clinical workflow.

Transcranial photoacoustic computed tomography (PACT) holds significant potential as a neuroimaging modality. However, compensating for skull-induced aberrations in reconstructed images remains a challenge. Although optimization-based image reconstruction methods (OBRMs) can account for the relevant wave physics, they are computationally demanding and generally require accurate estimates of the skull's viscoelastic parameters. To circumvent these issues, a learning-based image reconstruction method was investigated for three-dimensional (3D) transcranial PACT. The method was systematically assessed in virtual imaging studies that involved stochastic 3D numerical head phantoms and applied to experimental data acquired by use of a physical head phantom that involved a human skull. The results demonstrated that the learning-based method yielded accurate images and exhibited robustness to errors in the assumed skull properties, while substantially reducing computational times compared to an OBRM. To the best of our knowledge, this is the first demonstration of a learned image reconstruction method for 3D transcranial PACT.

To develop and validate an automated diabetic macular oedema (DME) classification system based on the images from different three-dimensional optical coherence tomography (3-D OCT) devices. A multicentre, platform-based development study using retrospective and cross-sectional data. Data were subjected to a two-level grading system by trained graders and a retina specialist, and categorised into three types: no DME, non-centre-involved DME and centre-involved DME (CI-DME). The 3-D convolutional neural networks algorithm was used for DME classification system development. The deep learning (DL) performance was compared with the diabetic retinopathy experts. Data were collected from Joint Shantou International Eye Center of Shantou University and the Chinese University of Hong Kong, Chaozhou People's Hospital and The Second Affiliated Hospital of Shantou University Medical College from January 2010 to December 2023. 7790 volumes of 7146 eyes from 4254 patients were annotated, of which 6281 images were used as the development set and 1509 images were used as the external validation set, split based on the centres. Accuracy, F1-score, sensitivity, specificity, area under receiver operating characteristic curve (AUROC) and Cohen's kappa were calculated to evaluate the performance of the DL algorithm. In classifying DME with non-DME, our model achieved an AUROCs of 0.990 (95% CI 0.983 to 0.996) and 0.916 (95% CI 0.902 to 0.930) for hold-out testing dataset and external validation dataset, respectively. To distinguish CI-DME from non-centre-involved-DME, our model achieved AUROCs of 0.859 (95% CI 0.812 to 0.906) and 0.881 (95% CI 0.859 to 0.902), respectively. In addition, our system showed comparable performance (Cohen's κ: 0.85 and 0.75) to the retina experts (Cohen's κ: 0.58-0.92 and 0.70-0.71). Our DL system achieved high accuracy in multiclassification tasks on DME classification with 3-D OCT images, which can be applied to population-based DME screening.

Dynamic Photoacoustic Computed Tomography (PACT) is an important imaging technique for monitoring physiological processes, capable of providing high-contrast images of optical absorption at much greater depths than traditional optical imaging methods. However, practical instrumentation and geometric constraints limit the number of acoustic sensors available around the imaging target, leading to sparsity in sensor data. Traditional photoacoustic (PA) image reconstruction methods, when directly applied to sparse PA data, produce severe artifacts. Additionally, these traditional methods do not consider the inter-frame relationships in dynamic imaging. Temporal resolution is crucial for dynamic photoacoustic imaging, which is fundamentally limited by the low repetition rate (e.g., 20 Hz) and high cost of high-power laser technology. Recently, Implicit Neural Representation (INR) has emerged as a powerful deep learning tool for solving inverse problems with sparse data, by characterizing signal properties as continuous functions of their coordinates in an unsupervised manner. In this work, we propose an INR-based method to improve dynamic photoacoustic image reconstruction from sparse-views and enhance temporal resolution, using only spatiotemporal coordinates as input. Specifically, the proposed INR represents dynamic photoacoustic images as implicit functions and encodes them into a neural network. The weights of the network are learned solely from the acquired sparse sensor data, without the need for external training datasets or prior images. Benefiting from the strong implicit continuity regularization provided by INR, as well as explicit regularization for low-rank and sparsity, our proposed method outperforms traditional reconstruction methods under two different sparsity conditions, effectively suppressing artifacts and ensuring image quality.

PurposeTo develop and validate deep learning (DL)-based models for classifying geographic atrophy (GA) subtypes using Optical Coherence Tomography (OCT) scans across four clinical classification tasks. DesignRetrospective comparative study evaluating three DL architectures on OCT data with two experimental approaches. Subjects455 OCT volumes (258 Central GA [CGA], 74 Non-Central GA [NCGA], 123 no GA [NGA]) from 104 patients at Atrium Health Wake Forest Baptist. For GA versus age-related macular degeneration (AMD) classification, we supplemented our dataset with AMD cases from four public repositories. MethodsWe implemented ResNet50, MobileNetV2, and Vision Transformer (ViT-B/16) architectures using two approaches: (1) utilizing all B-scans within each OCT volume and (2) selectively using B-scans containing foveal regions. Models were trained using transfer learning, standardized data augmentation, and patient-level data splitting (70:15:15 ratio) for training, validation, and testing. Main Outcome MeasuresArea under the receiver operating characteristic curve (AUC-ROC), F1 score, and accuracy for each classification task (CGA vs. NCGA, CGA vs. NCGA vs. NGA, GA vs. NGA, and GA vs. other forms of AMD). ResultsViT-B/16 consistently outperformed other architectures across all classification tasks. For CGA versus NCGA classification, ViT-B/16 achieved an AUC-ROC of 0.728{+/-}0.083 and accuracy of 0.831{+/-}0.006 using selective B-scans. In GA versus NGA classification, ViT-B/16 attained an AUC-ROC of 0.950{+/-}0.002 and accuracy of 0.873{+/-}0.012 with selective B-scans. All models demonstrated exceptional performance in distinguishing GA from other AMD forms (AUC-ROC>0.998). For multi-class classification, ViT-B/16 achieved an AUC-ROC of 0.873{+/-}0.003 and accuracy of 0.751{+/-}0.002 using selective B-scans. ConclusionsOur DL approach successfully classifies GA subtypes with clinically relevant accuracy. ViT-B/16 demonstrates superior performance due to its ability to capture spatial relationships between atrophic regions and the foveal center. Focusing on B-scans containing foveal regions improved diagnostic accuracy while reducing computational requirements, better aligning with clinical practice workflows.

Glaucoma is a group of serious eye diseases that can cause incurable blindness. Despite the critical need for early detection, over 60% of cases remain undiagnosed, especially in less developed regions. Glaucoma diagnosis is a costly task and some models have been proposed to automate diagnosis based on images of the retina, specifically the area known as the optic cup and the associated disc where retinal blood vessels and nerves enter and leave the eye. However, diagnosis is complicated because both normal and glaucoma-affected eyes can vary greatly in appearance. Some normal cases, like glaucoma, exhibit a larger cup-to-disc ratio, one of the main diagnostic criteria, making it challenging to distinguish between them. We propose a deep learning model with domain features (DLMDF) to combine unstructured and structured features to distinguish between glaucoma and physiologic large cups. The structured features were based upon the known cup-to-disc ratios of the four quadrants of the optic discs in normal, physiologic large cups, and glaucomatous optic cups. We segmented each cup and disc using a fully convolutional neural network and then calculated the cup size, disc size, and cup-to-disc ratio of each quadrant. The unstructured features were learned from a deep convolutional neural network. The average precision (AP) for disc segmentation was 98.52%, and for cup segmentation it was also 98.57%. Thus, the relatively high AP values enabled us to calculate the 15 reliable features from each segmented disc and cup. In classification tasks, the DLMDF outperformed other models, achieving superior accuracy, precision, and recall. These results validate the effectiveness of combining deep learning-derived features with domain-specific structured features, underscoring the potential of this approach to advance glaucoma diagnosis.

Breast cancer remains a leading cause of mortality among women worldwide, underscoring the need for accurate and timely diagnostic methods. Precise segmentation of nuclei in breast histopathology images is crucial for effective diagnosis and prognosis, offering critical insights into tumor characteristics and informing treatment strategies. This paper presents an enhanced U-Net architecture utilizing ResNet-34 as an advanced backbone, aimed at improving nuclei segmentation performance. The proposed model is evaluated and compared with standard U-Net and its other variants, including U-Net with VGG-16 and Inception-v3 backbones, using the BreCaHad dataset with nuclei masks generated through ImageJ software. The U-Net model with ResNet-34 backbone achieved superior performance, recording an Intersection over Union (IoU) score of 0.795, significantly outperforming the basic U-Net's IoU score of 0.725. The integration of advanced backbones and data augmentation techniques substantially improved segmentation accuracy, especially on limited medical imaging datasets. Comparative analysis demonstrated that ResNet-34 consistently surpassed other configurations across multiple metrics, including IoU, accuracy, precision, and F1 score. Further validation on the BNS and MoNuSeg-2018 datasets confirmed the robustness of the proposed model. This study highlights the potential of advanced deep learning architectures combined with augmentation methods to address challenges in nuclei segmentation, contributing to the development of more effective clinical diagnostic tools and improved patient care outcomes.

Exploring the trustworthiness of deep learning models is crucial, especially in critical domains such as medical imaging decision support systems. Conformal prediction has emerged as a rigorous means of providing deep learning models with reliable uncertainty estimates and safety guarantees. However, conformal prediction results face challenges due to the backbone model's struggles in domain-shifted scenarios, such as variations in different sources. To aim this challenge, this paper proposes a novel framework termed Conformal Ensemble of Vision Transformers (CE-ViTs) designed to enhance image classification performance by prioritizing domain adaptation and model robustness, while accounting for uncertainty. The proposed method leverages an ensemble of vision transformer models in the backbone, trained on diverse datasets including HAM10000, Dermofit, and Skin Cancer ISIC datasets. This ensemble learning approach, calibrated through the combined mentioned datasets, aims to enhance domain adaptation through conformal learning. Experimental results underscore that the framework achieves a high coverage rate of 90.38\%, representing an improvement of 9.95\% compared to the HAM10000 model. This indicates a strong likelihood that the prediction set includes the true label compared to singular models. Ensemble learning in CE-ViTs significantly improves conformal prediction performance, increasing the average prediction set size for challenging misclassified samples from 1.86 to 3.075.

To evaluate the impact of fluid volume fluctuations quantified with artificial intelligence in optical coherence tomography scans during the maintenance phase and visual outcomes at 12 and 24 months in a real-world, multicentre, national cohort of treatment-naïve neovascular age-related macular degeneration (nAMD) eyes. Demographics, visual acuity (VA) and number of injections were collected using the Fight Retinal Blindness tool. Intraretinal fluid (IRF), subretinal fluid (SRF), pigment epithelial detachment (PED), total fluid (TF) and central subfield thickness (CST) were quantified using the RetinAI Discovery tool. Fluctuations were defined as the SD of within-eye quantified values, and eyes were distributed according to SD quartiles for each biomarker. A total of 452 naïve nAMD eyes were included. Eyes with highest (Q4) versus lowest (Q1) fluid fluctuations showed significantly worse VA change (months 3-12) in IRF -3.91 versus 3.50 letters, PED -4.66 versus 3.29, TF -2.07 versus 2.97 and CST -1.85 versus 2.96 (all p<0.05), but not for SRF 0.66 versus 0.93 (p=0.91). Similar VA outcomes were observed at month 24 for PED -8.41 versus 4.98 (p<0.05), TF -7.38 versus 1.89 (p=0.07) and CST -10.58 versus 3.60 (p<0.05). The median number of injections (months 3-24) was significantly higher in Q4 versus Q1 eyes in IRF 9 versus 8, SRF 10 versus 8 and TF 10 versus 8 (all p<0.05). This multicentre study reports a negative effect in VA outcomes of fluid volume fluctuations during the maintenance phase in specific fluid compartments, suggesting that anatomical and functional treatment response patterns may be fluid-specific.