Purpose To develop and assess the accuracy of a hybrid deep learning construct for detecting keratoconus (KCN) based on corneal topographic maps. Methods We collected 3794 corneal images from 542 eyes of 280 subjects and developed seven deep learning models based on anterior and posterior eccentricity, anterior and posterior elevation, anterior and posterior sagittal curvature, and corneal thickness maps to extract deep corneal features. An independent subset with 1050 images collected from 150 eyes of 85 subjects from a separate center was used to validate models. We developed a hybrid deep learning model to detect KCN. We visualized deep features of corneal parameters to assess the quality of learning subjectively and computed area under the receiver operating characteristic curve (AUC), confusion matrices, accuracy, and F1 score to evaluate models objectively. Results In the development dataset, 204 eyes were normal, 123 eyes were suspected KCN, and 215 eyes had KCN. In the independent validation dataset, 50 eyes were normal, 50 eyes were suspected KCN, and 50 eyes were KCN. Images were annotated by three corneal specialists. The AUC of the models for the two-class and three-class problems based on the development set were 0.99 and 0.93, respectively. Conclusions The hybrid deep learning model achieved high accuracy in identifying KCN based on corneal maps and provided a time-efficient framework with low computational complexity. Translational Relevance Deep learning can detect KCN from non-invasive corneal images with high accuracy, suggesting potential application in research and clinical practice to identify KCN.

Clinical keratoconus (KCN) detection is a challenging and time-consuming task. In the diagnosis process, ophthalmologists must revise demographic and clinical ophthalmic examinations. The latter include slit-lamb, corneal topographic maps, and Pentacam indices (PI). We propose an Ensemble of Deep Transfer Learning (EDTL) based on corneal topographic maps. We consider four pretrained networks, SqueezeNet (SqN), AlexNet (AN), ShuffleNet (SfN), and MobileNet-v2 (MN), and fine-tune them on a dataset of KCN and normal cases, each including four topographic maps. We also consider a PI classifier. Then, our EDTL method combines the output probabilities of each of the five classifiers to obtain a decision based on the fusion of probabilities. Individually, the classifier based on PI achieved 93.1% accuracy, whereas the deep classifiers reached classification accuracies over 90% only in isolated cases. Overall, the average accuracy of the deep networks over the four corneal maps ranged from 86% (SfN) to 89.9% (AN). The classifier ensemble increased the accuracy of the deep classifiers based on corneal maps to values ranging (92.2% to 93.1%) for SqN and (93.1% to 94.8%) for AN. Including in the ensemble-specific combinations of corneal maps’ classifiers and PI increased the accuracy to 98.3%. Moreover, visualization of first learner filters in the networks and Grad-CAMs confirmed that the networks had learned relevant clinical features. This study shows the potential of creating ensembles of deep classifiers fine-tuned with a transfer learning strategy as it resulted in an improved accuracy while showing learnable filters and Grad-CAMs that agree with clinical knowledge. This is a step further towards the potential clinical deployment of an improved computer-assisted diagnosis system for KCN detection to help ophthalmologists to confirm the clinical decision and to perform fast and accurate KCN treatment.

Clinicians in the frontline need to assess quickly whether a patient with symptoms indeed has COVID-19 or not. The difficulty of this task is exacerbated in low resource settings that may not have access to biotechnology tests. Furthermore, Tuberculosis (TB) remains a major health problem in several lowand middle-income countries and its common symptoms include fever, cough and tiredness, similarly to COVID-19. In order to help in the detection of COVID-19, we propose the extraction of deep features (DF) from chest X-ray images, a technology available in most hospitals, and their subsequent classification using machine learning methods that do not require large computational resources. We compiled a five-class dataset of X-ray chest images including a balanced number of COVID-19, viral pneumonia, bacterial pneumonia, TB, and healthy cases. We compared the performance of pipelines combining 14 individual state-of-the-art pre-trained deep networks for DF extraction with traditional machine learning classifiers. A pipeline consisting of ResNet-50 for DF computation and ensemble of subspace discriminant classifier was the best performer in the classification of the five classes, achieving a detection accuracy of 91.6± 2.6% (accuracy ± Confidence Interval (CI) at 95% confidence level). Furthermore, the same pipeline achieved accuracies of 98.6±1.4% and 99.9±0.5% (± CI) in simpler three-class and two-class classification problems focused on distinguishing COVID-19, TB and healthy cases; and COVID-19 and healthy images, respectively. The pipeline was computationally efficient requiring just 0.19 second to extract DF per X-ray image and 2 minutes for training a traditional classifier with more than 2000 images on a CPU machine. The results suggest the potential benefits of using our pipeline in the detection of COVID-19, particularly in resource-limited settings as it relies in accessible X-rays and it can run with limited computational resources. The final constructed dataset named COVID-19 five-class balanced dataset is available from: https://drive.google.com/drive/folders/1toMymyHTy0DR_fyE7hjO3LSBGWtVoPNf?usp=sharing.

Detection of early clinical keratoconus (KCN) is a challenging task, even for expert clinicians. In this study, we propose a deep learning (DL) model to address this challenge. We first used Xception and InceptionResNetV2 DL architectures to extract features from three different corneal maps collected from 1371 eyes examined in an eye clinic in Egypt. We then fused features using Xception and InceptionResNetV2 to detect subclinical forms of KCN more accurately and robustly. We obtained an area under the receiver operating characteristic curves (AUC) of 0.99 and an accuracy range of 97–100% to distinguish normal eyes from eyes with subclinical and established KCN. We further validated the model based on an independent dataset with 213 eyes examined in Iraq and obtained AUCs of 0.91–0.92 and an accuracy range of 88–92%. The proposed model is a step toward improving the detection of clinical and subclinical forms of KCN.