Deep learning detection of melanoma metastases in lymph nodes

Computational methods for metastasis detection in lymph nodes and characterization of the metastasis-free lymph node microarchitecture: A systematic-narrative hybrid review

“…Jin et al 60 introduced ConcatNet model by concatenating four U-Net models each for nucleus, mitosis, epithelium, and tubule segmentation. Jansen et al 61 utilized U-Net model with ResNet50 encoder for metastasis identification in LNs from melanoma patients achieving competitive sensitivity scores of 91.67 and 95.62 on two datasets. Mainovskaya et al 62 used a pixel segmentation approach using DeepLabv3 model for metastasis segmentation in CRC LNs.…”

“…Based on the gathered data, LN sinus histiocytes were the most common false positively identified region mentioned by six studies 29 , 31 , 56 , 68 , 90 , 94 followed by secondary lymphoid follicles (germinal centers), 56 , 64 , 94 connective tissue, 31 , 64 out-of-focus areas, 68 and slide artifacts. 50 The most common metastatic regions misclassified as negative were micrometastatic lesions 61 , 64 as well as histiocyte-like tumor cells. 53 …”

“…Among patch classifiers, ResNet models were utilized most frequently (15 studies), 27 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 72 , 84 , 98 followed by MIL ( n =9), 31 , 41 , 77 , 78 , 79 , 80 , 81 , 82 , 99 VGG models ( n =8), 36 , 37 , 44 , 46 , 51 , 84 , 100 , 101 and others as shown in Fig. 4 B. Amongst pixel-level classification methods, U-Net was the most commonly used (10 studies) 26 , 27 , 30 , 59 , 60 , 61 , 62 , 71 , 96 , 102 followed by DeepLabv3 ( n =7), 28 , 29 , 62 , 63 , 64 , 65 , 66 Mask-RCNN ( n =1), 16 and proprietary Visiopharm AI metastasis model 90 as can be seen in Fig. 4 B.…”

Principles, applications, and future of artificial intelligence in dermatology

“…Aside from identifying the primary lesion, there are also studies exploring metastases. Jansen et al utilized histological tissue sections of sentinel lymph nodes in their convolutional neural network models to identify presence of metastases with high sensitivity and specificity ( 36 ).…”