Explore the vast range of titles available.

Machine learning techniques have great potential to improve medical diagnostics, offering ways to improve accuracy, reproducibility and speed, and to ease workloads for clinicians. In the field of histopathology, deep learning algorithms have been developed that perform similarly to trained pathologists for tasks such as tumor detection and grading. However, despite these promising results, very few algorithms have reached clinical implementation, challenging the balance between hope and hype for these new techniques. This Review provides an overview of the current state of the field, as well as describing the challenges that still need to be addressed before artificial intelligence in histopathology can achieve clinical value. Recent advances in machine learning techniques have created opportunities to improve medical diagnostics, but implementing these advances in the clinic will not be without challenge.

Objectives To investigate the added value of computer-aided diagnosis (CAD) on the diagnostic accuracy of PIRADS reporting and the assessment of cancer aggressiveness. Methods Multi-parametric MRI and histopathological outcome of MR-guided biopsies of a consecutive set of 130 patients were included. All cases were prospectively PIRADS reported and the reported lesions underwent CAD analysis. Logistic regression combined the CAD prediction and radiologist PIRADS score into a combination score. Receiver-operating characteristic (ROC) analysis and Spearman’s correlation coefficient were used to assess the diagnostic accuracy and correlation to cancer grade. Evaluation was performed for discriminating benign lesions from cancer and for discriminating indolent from aggressive lesions. Results In total 141 lesions (107 patients) were included for final analysis. The area-under-the-ROC-curve of the combination score was higher than for the PIRADS score of the radiologist (benign vs. cancer, 0.88 vs. 0.81, p = 0.013 and indolent vs. aggressive, 0.88 vs. 0.78, p < 0.01). The combination score correlated significantly stronger with cancer grade (0.69, p = 0.0014) than the individual CAD system or radiologist (0.54 and 0.58). Conclusions Combining CAD prediction and PIRADS into a combination score has the potential to improve diagnostic accuracy. Furthermore, such a combination score has a strong correlation with cancer grade. Key Points • Computer-aided diagnosis helps radiologists discriminate benign findings from cancer in prostate MRI. • Combining PIRADS and computer-aided diagnosis improves differentiation between indolent and aggressive cancer. • Adding computer-aided diagnosis to PIRADS increases the correlation coefficient with respect to cancer grade.

The Gleason score is the strongest correlating predictor of recurrence for prostate cancer, but has substantial inter-observer variability, limiting its usefulness for individual patients. Specialised urological pathologists have greater concordance; however, such expertise is not widely available. Prostate cancer diagnostics could thus benefit from robust, reproducible Gleason grading. We aimed to investigate the potential of deep learning to perform automated Gleason grading of prostate biopsies. In this retrospective study, we developed a deep-learning system to grade prostate biopsies following the Gleason grading standard. The system was developed using randomly selected biopsies, sampled by the biopsy Gleason score, from patients at the Radboud University Medical Center (pathology report dated between Jan 1, 2012, and Dec 31, 2017). A semi-automatic labelling technique was used to circumvent the need for manual annotations by pathologists, using pathologists' reports as the reference standard during training. The system was developed to delineate individual glands, assign Gleason growth patterns, and determine the biopsy-level grade. For validation of the method, a consensus reference standard was set by three expert urological pathologists on an independent test set of 550 biopsies. Of these 550, 100 were used in an observer experiment, in which the system, 13 pathologists, and two pathologists in training were compared with respect to the reference standard. The system was also compared to an external test dataset of 886 cores, which contained 245 cores from a different centre that were independently graded by two pathologists. We collected 5759 biopsies from 1243 patients. The developed system achieved a high agreement with the reference standard (quadratic Cohen's kappa 0·918, 95% CI 0·891–0·941) and scored highly at clinical decision thresholds: benign versus malignant (area under the curve 0·990, 95% CI 0·982–0·996), grade group of 2 or more (0·978, 0·966–0·988), and grade group of 3 or more (0·974, 0·962–0·984). In an observer experiment, the deep-learning system scored higher (kappa 0·854) than the panel (median kappa 0·819), outperforming 10 of 15 pathologist observers. On the external test dataset, the system obtained a high agreement with the reference standard set independently by two pathologists (quadratic Cohen's kappa 0·723 and 0·707) and within inter-observer variability (kappa 0·71). Our automated deep-learning system achieved a performance similar to pathologists for Gleason grading and could potentially contribute to prostate cancer diagnosis. The system could potentially assist pathologists by screening biopsies, providing second opinions on grade group, and presenting quantitative measurements of volume percentages. Dutch Cancer Society.

International challenges have become the de facto standard for comparative assessment of image analysis algorithms. Although segmentation is the most widely investigated medical image processing task, the various challenges have been organized to focus only on specific clinical tasks. We organized the Medical Segmentation Decathlon (MSD)—a biomedical image analysis challenge, in which algorithms compete in a multitude of both tasks and modalities to investigate the hypothesis that a method capable of performing well on multiple tasks will generalize well to a previously unseen task and potentially outperform a custom-designed solution. MSD results confirmed this hypothesis, moreover, MSD winner continued generalizing well to a wide range of other clinical problems for the next two years. Three main conclusions can be drawn from this study: (1) state-of-the-art image segmentation algorithms generalize well when retrained on unseen tasks; (2) consistent algorithmic performance across multiple tasks is a strong surrogate of algorithmic generalizability; (3) the training of accurate AI segmentation models is now commoditized to scientists that are not versed in AI model training. International challenges have become the de facto standard for comparative assessment of image analysis algorithms. Here, the authors present the results of a biomedical image segmentation challenge, showing that a method capable of performing well on multiple tasks will generalize well to a previously unseen task.

Algorithms can improve the objectivity and efficiency of histopathologic slide analysis. In this paper, we investigated the impact of scanning systems (scanners) and cycle-GAN-based normalization on algorithm performance, by comparing different deep learning models to automatically detect prostate cancer in whole-slide images. Specifically, we compare U-Net, DenseNet and EfficientNet. Models were developed on a multi-center cohort with 582 WSIs and subsequently evaluated on two independent test sets including 85 and 50 WSIs, respectively, to show the robustness of the proposed method to differing staining protocols and scanner types. We also investigated the application of normalization as a pre-processing step by two techniques, the whole-slide image color standardizer (WSICS) algorithm, and a cycle-GAN based method. For the two independent datasets we obtained an AUC of 0.92 and 0.83 respectively. After rescanning the AUC improves to 0.91/0.88 and after style normalization to 0.98/0.97. In the future our algorithm could be used to automatically pre-screen prostate biopsies to alleviate the workload of pathologists.

To investigate whether multi-view convolutional neural networks can improve a fully automated lymph node detection system for pelvic MR Lymphography (MRL) images of patients with prostate cancer. A fully automated computer-aided detection (CAD) system had been previously developed to detect lymph nodes in MRL studies. The CAD system was extended with three types of 2D multi-view convolutional neural networks (CNN) aiming to reduce false positives (FP). A 2D multi-view CNN is an efficient approximation of a 3D CNN, and three types were evaluated: a 1-view, 3-view, and 9-view 2D CNN. The three deep learning CNN architectures were trained and configured on retrospective data of 240 prostate cancer patients that received MRL images as the standard of care between January 2008 and April 2010. The MRL used ferumoxtran-10 as a contrast agent and comprised at least two imaging sequences: a 3D T1-weighted and a 3D T2*-weighted sequence. A total of 5089 lymph nodes were annotated by two expert readers, reading in consensus. A first experiment compared the performance with and without CNNs and a second experiment compared the individual contribution of the 1-view, 3-view, or 9-view architecture to the performance. The performances were visually compared using free-receiver operating characteristic (FROC) analysis and statistically compared using partial area under the FROC curve analysis. Training and analysis were performed using bootstrapped FROC and 5-fold cross-validation. Adding multi-view CNNs significantly (  < 0.01) reduced false positive detections. The 3-view and 9-view CNN outperformed (  < 0.01) the 1-view CNN, reducing FP from 20.6 to 7.8/image at 80% sensitivity. Multi-view convolutional neural networks significantly reduce false positives in a lymph node detection system for MRL images, and three orthogonal views are sufficient. At the achieved level of performance, CAD for MRL may help speed up finding lymph nodes and assessing them for potential metastatic involvement.

Artificial intelligence (AI) has shown promise for diagnosing prostate cancer in biopsies. However, results have been limited to individual studies, lacking validation in multinational settings. Competitions have been shown to be accelerators for medical imaging innovations, but their impact is hindered by lack of reproducibility and independent validation. With this in mind, we organized the PANDA challenge—the largest histopathology competition to date, joined by 1,290 developers—to catalyze development of reproducible AI algorithms for Gleason grading using 10,616 digitized prostate biopsies. We validated that a diverse set of submitted algorithms reached pathologist-level performance on independent cross-continental cohorts, fully blinded to the algorithm developers. On United States and European external validation sets, the algorithms achieved agreements of 0.862 (quadratically weighted κ, 95% confidence interval (CI), 0.840–0.884) and 0.868 (95% CI, 0.835–0.900) with expert uropathologists. Successful generalization across different patient populations, laboratories and reference standards, achieved by a variety of algorithmic approaches, warrants evaluating AI-based Gleason grading in prospective clinical trials. Through a community-driven competition, the PANDA challenge provides a curated diverse dataset and a catalog of models for prostate cancer pathology, and represents a blueprint for evaluating AI algorithms in digital pathology.

Pathologists face a substantial increase in workload and complexity of histopathologic cancer diagnosis due to the advent of personalized medicine. Therefore, diagnostic protocols have to focus equally on efficiency and accuracy. In this paper we introduce ‘deep learning’ as a technique to improve the objectivity and efficiency of histopathologic slide analysis. Through two examples, prostate cancer identification in biopsy specimens and breast cancer metastasis detection in sentinel lymph nodes, we show the potential of this new methodology to reduce the workload for pathologists, while at the same time increasing objectivity of diagnoses. We found that all slides containing prostate cancer and micro- and macro-metastases of breast cancer could be identified automatically while 30–40% of the slides containing benign and normal tissue could be excluded without the use of any additional immunohistochemical markers or human intervention. We conclude that ‘deep learning’ holds great promise to improve the efficacy of prostate cancer diagnosis and breast cancer staging.

In patients with suspected lymphoma, the tissue biopsy provides lymphoma confirmation, classification, and prognostic factors, including genetic changes. We developed a deep learning algorithm to detect MYC rearrangement in scanned histological slides of diffuse large B-cell lymphoma. The H&E-stained slides of 287 cases from 11 hospitals were used for training and evaluation. The overall sensitivity to detect MYC rearrangement was 0.93 and the specificity 0.52, showing that prediction of MYC translocation based on morphology alone was possible in 93% of MYC-rearranged cases. This would allow a simple and fast prescreening, saving approximately 34% of genetic tests with the current algorithm.

The anatomical location of imaging features is of crucial importance for accurate diagnosis in many medical tasks. Convolutional neural networks (CNN) have had huge successes in computer vision, but they lack the natural ability to incorporate the anatomical location in their decision making process, hindering success in some medical image analysis tasks. In this paper, to integrate the anatomical location information into the network, we propose several deep CNN architectures that consider multi-scale patches or take explicit location features while training. We apply and compare the proposed architectures for segmentation of white matter hyperintensities in brain MR images on a large dataset. As a result, we observe that the CNNs that incorporate location information substantially outperform a conventional segmentation method with handcrafted features as well as CNNs that do not integrate location information. On a test set of 50 scans, the best configuration of our networks obtained a Dice score of 0.792, compared to 0.805 for an independent human observer. Performance levels of the machine and the independent human observer were not statistically significantly different (p-value = 0.06).