No abstract

Deep learning‐based computer vision methods have recently made remarkable breakthroughs in the analysis and classification of cancer pathology images. However, there has been relatively little investigation of the utility of deep neural networks to synthesize medical images. In this study, we evaluated the efficacy of generative adversarial networks to synthesize high‐resolution pathology images of 10 histological types of cancer, including five cancer types from The Cancer Genome Atlas and the five major histological subtypes of ovarian carcinoma. The quality of these images was assessed using a comprehensive survey of board‐certified pathologists (n = 9) and pathology trainees (n = 6). Our results show that the real and synthetic images are classified by histotype with comparable accuracies and the synthetic images are visually indistinguishable from real images. Furthermore, we trained deep convolutional neural networks to diagnose the different cancer types and determined that the synthetic images perform as well as additional real images when used to supplement a small training set. These findings have important applications in proficiency testing of medical practitioners and quality assurance in clinical laboratories. Furthermore, training of computer‐aided diagnostic systems can benefit from synthetic images where labeled datasets are limited (e.g. rare cancers). We have created a publicly available website where clinicians and researchers can attempt questions from the image survey (http://gan.aimlab.ca/). © 2020 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Background: Spinal cord stimulation (SCS) is a well-established treatment for chronic neuropathic pain in the lower limbs. Upper limb pain comprises a significant proportion of neuropathic pain patients, but is often difficult to target specifically and consistently with paresthesias. We hypothesized that the use of dorsal nerve root stimulation (DNRS), as an option along with SCS, would help us better relieve pain in these patients. Methods: All 35 patients trialed with spinal stimulation for upper limb pain between July 1, 2011, and October 31, 2013, were included. We performed permanent implantation in 23/35 patients based on a visual analogue scale pain score decrease of ≥50% during trial stimulation. Results: Both the SCS and DNRS groups had significant improvements in average visual analogue scale pain scores at 12 months compared with baseline, and the majority of patients in both groups obtained ≥50% pain relief. The majority of patients in both groups were able to reduce their opioid use, and on average had improvements in Short Form-36 quality of life scores. Complication rates did not differ significantly between the two groups. Conclusions: Treatment with SCS or DNRS provides meaningful long-term relief of chronic neuropathic pain in the upper limbs.RÉSUMÉ: Stimulation de la moelle épinière et de la racine dorsale pour soulager la douleur neurogène des membres supérieurs. Contexte: La stimulation de la moelle épinière (SME) est un traitement de la douleur neurogène chronique des membres inférieurs qui a fait ses preuves. Une proportion importante de patients souffre aussi de douleur neurogène aux membres supérieurs mais il demeure ardu de cibler une telle douleur de façon systématique et spécifique en lien avec des manifestations de paresthésie. Nous avons ainsi formulé l'hypothèse que la stimulation de la racine dorsale, en plus de la SME, pourrait nous aider à mieux soulager la douleur chez ces patients. Méthodes: Tous les 35 patients chez qui on avait effectué, du 1 er juillet 2011 au 31 octobre 2013, un traitement de SME des membres supérieurs ont été inclus dans cette étude. Pendant les essais cliniques de stimulation, nous avons soumis 23 patients sur 35 à un traitement continu. En nous fondant sur l'échelle visuelle analogique (ÉVA), nous avons anticipé une réduction de ≥50% du score lié à la douleur. Résultats: Au bout de 12 mois, tant les groupes ayant bénéficié de la SME que ceux ayant bénéficié de la stimulation de la racine dorsale ont obtenu, par rapport à des valeurs de référence, des scores nettement meilleurs en matière d'ÉVA de la douleur. En effet, une majorité de patients des deux groupes a rapporté un soulagement de la douleur de ≥50%. Une majorité d'entre eux a aussi été en mesure de réduire sa consommation d'opiacés et a amélioré son score au test SF-36 (Short Form 36 Health Survey) en matière de qualité de vie. Fait à souligner, il n'y a pas eu de différence notable entre les deux groupes quant aux taux de complication. Conclusions: Tant la SME que la stimulation de l...

The color variation of hematoxylin and eosin (H&E)-stained tissues has presented a challenge for applications of artificial intelligence (AI) in digital pathology. Many color normalization algorithms have been developed in recent years in order to reduce the color variation between H&E images. However, previous efforts in benchmarking these algorithms have produced conflicting results and none have sufficiently assessed the efficacy of the various color normalization methods for improving diagnostic performance of AI systems. In this study, we systematically investigated eight color normalization algorithms for AI-based classification of H&E-stained histopathology slides, in the context of using images both from one center and from multiple centers. Our results show that color normalization does not consistently improve classification performance when both training and testing data are from a single center. However, using four multi-center datasets of two cancer types (ovarian and pleural) and objective functions, we show that color normalization can significantly improve the classification accuracy of images from external datasets (ovarian cancer: 0.25 AUC increase, p = 1.6 e-05; pleural cancer: 0.21 AUC increase, p = 1.4 e-10). Furthermore, we introduce a novel augmentation strategy by mixing color-normalized images using three easily accessible algorithms that consistently improves the diagnosis of test images from external centers, even when the individual normalization methods had varied results. We anticipate our study to be a starting point for reliable use of color normalization to improve AI-based, digital pathology-empowered diagnosis of cancers sourced from multiple centers.