Multiscale Unified Network for Simultaneous Segmentation of Nerves and Micro-vessels in Histology Images

Abstract

Among the analytic factors to study tumor aggressiveness and disease recurrence, density of micro-vessels (MVD), Lymphovascular Invasion (LVI) and Perineural Invasion (PNI) are considered key prognostic factors. The manual identification of micro-vessels and nerves is time consuming, laborious and highly prone to human error. Computational pathology is an emerging field striving to improve patient care by incorporating modern algorithms to the traditional analysis procedures of microscopic slides. To overcome the challenges of multi-scale, multi-shape and slight intensity variant histopathology structures, we have proposed a deep neural network based hybrid semantic segmentation architecture. The framework is specifically designed to improve the accuracy by focusing mega to minor object details. The encoder uses Multi-scale feature extraction block formed of ResNeXt Blocks. This organization is effective to encode coarse to fine grained features from all specifications and dimensions while limiting the number of learnable parameters. The decoder is a combination of feature fusion and feature erudition while step by step mapping them back to the pixel map. The proposed architecture is trained and tested on generated data set comprising 17,300 samples, prepared from 18 histopathological WSIs of oral cell carcinoma tissues. The trained architecture outperformed the existing segmentation networks like FCN, Unet, SegNet, Deeplabv3+ and a significant rise in accuracy regarding certain scenarios is observed.