RESUMO
Even though artificial intelligence and machine learning have demonstrated remarkable performances in medical image computing, their accountability and transparency level must be improved to transfer this success into clinical practice. The reliability of machine learning decisions must be explained and interpreted, especially for supporting the medical diagnosis. For this task, the deep learning techniques' black-box nature must somehow be lightened up to clarify its promising results. Hence, we aim to investigate the impact of the ResNet-50 deep convolutional design for Barrett's esophagus and adenocarcinoma classification. For such a task, and aiming at proposing a two-step learning technique, the output of each convolutional layer that composes the ResNet-50 architecture was trained and classified for further definition of layers that would provide more impact in the architecture. We showed that local information and high-dimensional features are essential to improve the classification for our task. Besides, we observed a significant improvement when the most discriminative layers expressed more impact in the training and classification of ResNet-50 for Barrett's esophagus and adenocarcinoma classification, demonstrating that both human knowledge and computational processing may influence the correct learning of such a problem.
RESUMO
BACKGROUND: This study evaluated the effect of an artificial intelligence (AI)-based clinical decision support system on the performance and diagnostic confidence of endoscopists in their assessment of Barrett's esophagus (BE). METHODS: 96 standardized endoscopy videos were assessed by 22 endoscopists with varying degrees of BE experience from 12 centers. Assessment was randomized into two video sets: group A (review first without AI and second with AI) and group B (review first with AI and second without AI). Endoscopists were required to evaluate each video for the presence of Barrett's esophagus-related neoplasia (BERN) and then decide on a spot for a targeted biopsy. After the second assessment, they were allowed to change their clinical decision and confidence level. RESULTS: AI had a stand-alone sensitivity, specificity, and accuracy of 92.2%, 68.9%, and 81.3%, respectively. Without AI, BE experts had an overall sensitivity, specificity, and accuracy of 83.3%, 58.1%, and 71.5%, respectively. With AI, BE nonexperts showed a significant improvement in sensitivity and specificity when videos were assessed a second time with AI (sensitivity 69.8% [95%CI 65.2%-74.2%] to 78.0% [95%CI 74.0%-82.0%]; specificity 67.3% [95%CI 62.5%-72.2%] to 72.7% [95%CI 68.2%-77.3%]). In addition, the diagnostic confidence of BE nonexperts improved significantly with AI. CONCLUSION: BE nonexperts benefitted significantly from additional AI. BE experts and nonexperts remained significantly below the stand-alone performance of AI, suggesting that there may be other factors influencing endoscopists' decisions to follow or discard AI advice.
RESUMO
Semantic segmentation is an essential task in medical imaging research. Many powerful deep-learning-based approaches can be employed for this problem, but they are dependent on the availability of an expansive labeled dataset. In this work, we augment such supervised segmentation models to be suitable for learning from unlabeled data. Our semi-supervised approach, termed Error-Correcting Mean-Teacher, uses an exponential moving average model like the original Mean Teacher but introduces our new paradigm of error correction. The original segmentation network is augmented to handle this secondary correction task. Both tasks build upon the core feature extraction layers of the model. For the correction task, features detected in the input image are fused with features detected in the predicted segmentation and further processed with task-specific decoder layers. The combination of image and segmentation features allows the model to correct present mistakes in the given input pair. The correction task is trained jointly on the labeled data. On unlabeled data, the exponential moving average of the original network corrects the student's prediction. The combined outputs of the students' prediction with the teachers' correction form the basis for the semi-supervised update. We evaluate our method with the 2017 and 2018 Robotic Scene Segmentation data, the ISIC 2017 and the BraTS 2020 Challenges, a proprietary Endoscopic Submucosal Dissection dataset, Cityscapes, and Pascal VOC 2012. Additionally, we analyze the impact of the individual components and examine the behavior when the amount of labeled data varies, with experiments performed on two distinct segmentation architectures. Our method shows improvements in terms of the mean Intersection over Union over the supervised baseline and competing methods. Code is available at https://github.com/CloneRob/ECMT.
Assuntos
Pesquisa Biomédica , Robótica , Humanos , Semântica , Processamento de Imagem Assistida por ComputadorRESUMO
BACKGROUND AND AIMS: Celiac disease with its endoscopic manifestation of villous atrophy (VA) is underdiagnosed worldwide. The application of artificial intelligence (AI) for the macroscopic detection of VA at routine EGD may improve diagnostic performance. METHODS: A dataset of 858 endoscopic images of 182 patients with VA and 846 images from 323 patients with normal duodenal mucosa was collected and used to train a ResNet18 deep learning model to detect VA. An external dataset was used to test the algorithm, in addition to 6 fellows and 4 board-certified gastroenterologists. Fellows could consult the AI algorithm's result during the test. From their consultation distribution, a stratification of test images into "easy" and "difficult" was performed and used for classified performance measurement. RESULTS: External validation of the AI algorithm yielded values of 90%, 76%, and 84% for sensitivity, specificity, and accuracy, respectively. Fellows scored corresponding values of 63%, 72%, and 67% and experts scored 72%, 69%, and 71%, respectively. AI consultation significantly improved all trainee performance statistics. Although fellows and experts showed significantly lower performance for difficult images, the performance of the AI algorithm was stable. CONCLUSIONS: In this study, an AI algorithm outperformed endoscopy fellows and experts in the detection of VA on endoscopic still images. AI decision support significantly improved the performance of nonexpert endoscopists. The stable performance on difficult images suggests a further positive add-on effect in challenging cases.
Assuntos
Inteligência Artificial , Aprendizado Profundo , Humanos , Endoscopia Gastrointestinal , Algoritmos , AtrofiaRESUMO
In this study, we aimed to develop an artificial intelligence clinical decision support solution to mitigate operator-dependent limitations during complex endoscopic procedures such as endoscopic submucosal dissection and peroral endoscopic myotomy, for example, bleeding and perforation. A DeepLabv3-based model was trained to delineate vessels, tissue structures and instruments on endoscopic still images from such procedures. The mean cross-validated Intersection over Union and Dice Score were 63% and 76%, respectively. Applied to standardised video clips from third-space endoscopic procedures, the algorithm showed a mean vessel detection rate of 85% with a false-positive rate of 0.75/min. These performance statistics suggest a potential clinical benefit for procedure safety, time and also training.
Assuntos
Aprendizado Profundo , Ressecção Endoscópica de Mucosa , Humanos , Inteligência Artificial , Endoscopia GastrointestinalRESUMO
The endoscopic features associated with eosinophilic esophagitis (EoE) may be missed during routine endoscopy. We aimed to develop and evaluate an Artificial Intelligence (AI) algorithm for detecting and quantifying the endoscopic features of EoE in white light images, supplemented by the EoE Endoscopic Reference Score (EREFS). An AI algorithm (AI-EoE) was constructed and trained to differentiate between EoE and normal esophagus using endoscopic white light images extracted from the database of the University Hospital Augsburg. In addition to binary classification, a second algorithm was trained with specific auxiliary branches for each EREFS feature (AI-EoE-EREFS). The AI algorithms were evaluated on an external data set from the University of North Carolina, Chapel Hill (UNC), and compared with the performance of human endoscopists with varying levels of experience. The overall sensitivity, specificity, and accuracy of AI-EoE were 0.93 for all measures, while the AUC was 0.986. With additional auxiliary branches for the EREFS categories, the AI algorithm (AI-EoE-EREFS) performance improved to 0.96, 0.94, 0.95, and 0.992 for sensitivity, specificity, accuracy, and AUC, respectively. AI-EoE and AI-EoE-EREFS performed significantly better than endoscopy beginners and senior fellows on the same set of images. An AI algorithm can be trained to detect and quantify endoscopic features of EoE with excellent performance scores. The addition of the EREFS criteria improved the performance of the AI algorithm, which performed significantly better than endoscopists with a lower or medium experience level.
Assuntos
Esofagite Eosinofílica , Inteligência Artificial , Esofagite Eosinofílica/diagnóstico , Esofagoscopia/métodos , Humanos , Índice de Gravidade de DoençaRESUMO
BACKGROUND: The accurate differentiation between T1a and T1b Barrett's-related cancer has both therapeutic and prognostic implications but is challenging even for experienced physicians. We trained an artificial intelligence (AI) system on the basis of deep artificial neural networks (deep learning) to differentiate between T1a and T1b Barrett's cancer on white-light images. METHODS: Endoscopic images from three tertiary care centers in Germany were collected retrospectively. A deep learning system was trained and tested using the principles of cross validation. A total of 230 white-light endoscopic images (108 T1a and 122 T1b) were evaluated using the AI system. For comparison, the images were also classified by experts specialized in endoscopic diagnosis and treatment of Barrett's cancer. RESULTS: The sensitivity, specificity, F1 score, and accuracy of the AI system in the differentiation between T1a and T1b cancer lesions was 0.77, 0.64, 0.74, and 0.71, respectively. There was no statistically significant difference between the performance of the AI system and that of experts, who showed sensitivity, specificity, F1, and accuracy of 0.63, 0.78, 0.67, and 0.70, respectively. CONCLUSION: This pilot study demonstrates the first multicenter application of an AI-based system in the prediction of submucosal invasion in endoscopic images of Barrett's cancer. AI scored equally to international experts in the field, but more work is necessary to improve the system and apply it to video sequences and real-life settings. Nevertheless, the correct prediction of submucosal invasion in Barrett's cancer remains challenging for both experts and AI.