Portal de Periódicos da CAPES

Tumor Infiltrating Lymphocytes Recognition in Primary Melanoma by Deep Learning Convolutional Neuronal Network

2022; RELX Group (Netherlands); Linguagem: Inglês

10.2139/ssrn.4233979

1556-5068

Filippo Ugolini, Francesco De Logu, Luigi Francesco Iannone, Francesca Brutti, Sara Simi, Vincenza Maio, Vincenzo De Giorgi, Anna Maria Di Giacomo, Clelia Miracco, Francesco Federico, Ketty Peris, Giuseppe Palmieri, Antonio Cossu, Mario Mandalà, Daniela Massi, Marco Laurino,

Cutaneous Melanoma Detection and Management

The presence of tumor-infiltrating lymphocytes (TIL) has been associated with a favorable prognosis of primary melanoma (PM). The recent development of the artificial intelligence (AI) based approach in digital pathology has been proposed for the standardized assessment of TIL on hematoxylin and eosin (H&E)-stained images (whole slide images, WSI). Here, we have applied a new convolution neural network (CNN) analysis of PM WSI to automatically assess the infiltration of TILs and extract a TILs score. A CNN based on a pre-trained Inception-ResNet-v2 was trained and validated in a retrospective cohort of 307 PMs including a training set (N = 237 WSI for 57,758 patches) and an independent testing set (N = 70 WSI for 29,533 patches). After the classification of tumor patches by the presence or absence of TILs, we identified an AI-based TILs density index (AI-TIL). The proposed CNN demonstrated high performance in recognizing TILs in PM WSI, showing specificity and sensitivity of 100% on the testing set. We demonstrated that the AI-based TILs index correlated with conventional TILs evaluation and clinical outcome. The present AI-TIL index was an independent prognostic marker directly associated with a favorable prognosis. A fully automated and standardized AI-TIL appears to be superior to conventional methods at differentiating PM clinical outcome. Further studies are required to develop an easy-to-use tool that could assist pathologists to assess TILs in the clinical evaluation of solid tumors.