Ophthalmological services face global inadequacies, especially in low- and middle-income countries, which are marked by a shortage of practitioners and equipment. This study employed a portable slit lamp microscope with video capabilities and cloud storage for more equitable global diagnostic resource distribution. To enhance accessibility and quality of care, this study targets corneal opacity, which is a global cause of blindness. This study has two purposes. The first is to detect corneal opacity from videos in which the anterior segment of the eye is captured. The other is to develop an AI pipeline to detect corneal opacities. First, we extracted image frames from videos and processed them using a convolutional neural network (CNN) model. Second, we manually annotated the images to extract only the corneal margins, adjusted the contrast with CLAHE, and processed them using the CNN model. Finally, we performed semantic segmentation of the cornea using annotated data. The results showed an accuracy of 0.8 for image frames and 0.96 for corneal margins. Dice and IoU achieved a score of 0.94 for semantic segmentation of the corneal margins. Although corneal opacity detection from video frames seemed challenging in the early stages of this study, manual annotation, corneal extraction, and CLAHE contrast adjustment significantly improved accuracy. The incorporation of manual annotation into the AI pipeline, through semantic segmentation, facilitated high accuracy in detecting corneal opacity.

Deep learning has achieved widespread adoption for medical image diagnosis, with extensive research dedicated to mammographic image analysis for breast cancer screening. This study investigates the hypothesis that incorporating region-of-interest (ROI) mask information for individual mammographic images during deep learning can improve the accuracy of benign/malignant diagnoses. Swin Transformer and ConvNeXtV2 deep learning models were used to evaluate their performance on the public VinDr and CDD-CESM datasets. Our approach involved stratifying mammographic images based on the presence or absence of ROI masks, performing independent training and prediction for each subgroup, and subsequently merging the results. Baseline prediction metrics (sensitivity, specificity, F-score, and accuracy) without ROI-stratified separation were the following: VinDr/Swin Transformer (0.00, 1.00, 0.00, 0.85), VinDr/ConvNeXtV2 (0.00, 1.00, 0.00, 0.85), CDD-CESM/Swin Transformer (0.29, 0.68, 0.41, 0.48), and CDD-CESM/ConvNeXtV2 (0.65, 0.65, 0.65, 0.65). Subsequent analysis with ROI-stratified separation demonstrated marked improvements in these metrics: VinDr/Swin Transformer (0.93, 0.87, 0.90, 0.87), VinDr/ConvNeXtV2 (0.90, 0.86, 0.88, 0.87), CDD-CESM/Swin Transformer (0.65, 0.65, 0.65, 0.65), and CDD-CESM/ConvNeXtV2 (0.74, 0.61, 0.67, 0.68). These findings provide compelling evidence that validate our hypothesis and affirm the utility of considering ROI mask information for enhanced diagnostic accuracy in mammography.

Ophthalmological services face global inadequacies, especially in low- and middle-income countries, which are marked by a shortage of both practitioners and equipment. This study employed a portable slit-lamp microscope with video capabilities and cloud storage for a more equitable global diagnostic resource distribution. To enhance accessibility and quality of care, this study targeted corneal opacity, a global cause of blindness. To boost the online diagnosis efficiency, an AI pipeline was developed using anterior segment videos to detect corneal opacity. First, we extracted image frames from videos and learned them using a Convolutional Neural Network(CNN) model. Second, we manually annotated to extract only the corneal margins, adjusted the contrast with CLAHE, and learned using the CNN model. Finally, we performed semantic segmentation of the cornea using the annotated data. The results showed an accuracy of 0.8 for image frames, and 0.96 for corneal margins. Dice and IoU were 0.94 for semantic segmentation of corneal margins. While corneal opacity detection from video frames seemed challenging in the early stages of this study, manual annotation, corneal extraction, and CLAHE contrast adjustment significantly improved accuracy. Integrating manual annotation into the AI pipeline through semantic segmentation achieved a high accuracy in detecting corneal opacity.

Deep Learning (DL) has undergone widespread adoption for medical image analysis and diagnosis. Numerous studies have explored mammographic image analysis for breast cancer screening. For this study, we assessed the hypothesis that stratifying mammography images based on the presence or absence of a corresponding region of interest (ROI) improves classification accuracy for both normal–abnormal and benign–malignant classifications. Our methodology involves independently training models and performing predictions on each subgroup with subsequent integration of the results. We used several DL models, including ResNet, EfficientNet, SwinTransformer, ConvNeXt, and MobileNet. For experimentation, we used the publicly available VinDr., CDD-CESM, and DMID datasets. Our comparison with prediction results obtained without ROI-based stratification demonstrated that the utility of considering ROI presence to enhance diagnostic accuracy in mammography increases along with the data volume. These findings support the usefulness of our stratification approach, particularly as a dataset size grows.

Deep Learning (DL) has undergone widespread adoption for medical image analysis and diagnosis. Numerous studies have explored mammographic image analysis for breast cancer screening. For this study, we assessed the hypothesis that stratifying mammography images based on the presence or absence of a corresponding region of interest (ROI) improves classification accuracy for both normal–abnormal and benign–malignant classifications. Our methodology involves independently training models and performing predictions on each subgroup with subsequent integration of the results. We used several DL models, including ResNet, EfficientNet, SwinTransformer, ConvNeXt, and MobileNet. For experimentation, we used the publicly available VinDr., CDD-CESM, and DMID datasets. Our comparison with prediction results obtained without ROI-based stratification demonstrated that the utility of considering ROI presence to enhance diagnostic accuracy in mammography increases along with the data volume. These findings support the usefulness of our stratification approach, particularly as a dataset’s size grows.

Deep learning is currently being used for image diagnosis in medicine. If image diagnosis can be performed with high sensitivity using deep learning, it will be possible to reduce the burden on medical specialists reading images. However, abnormal medical images are often difficult to detect by deep learning due to the large number of normal medical images. To improve such unbalanced classification sensitivity, we improved the sensitivity of predicting whether a medical image is benign or malignant by up-sampling the numerical features obtained by radiomics from the image and randomly selecting and averaging them. The unbalanced VinDr-Mammo dataset, with 988 malignant and 5606 benign cases, was used for our study. The sensitivity for deep learning image classification was 0.2626 for CNN, 0.0000 for VisionTransformer, 0.0101 for Radom Forest (radiomics features), 0.0101 for Radom Forest (upscaled radiomics features) was 0.4141. The sensitivity for the ensemble of predictions by CNN and Radom Forest (upscaled radiomics features) was 0.5303. The results suggest the effectiveness of the up-sampled radiomics feature method, which randomly selects and averages features in an unbalanced medical image classification. This research leaves room for investigation using sampling methods such as SMOTE and AdaBoost, as well as research into the robustness and generalizability of the model to diverse datasets.