Instant diagnosis of gastroscopic biopsy via deep-learned single-shot femtosecond stimulated Raman histology

Abstract

Gastroscopic biopsy provides the only effective method for gastric cancer diagnosis, but the gold standard histopathology is time-consuming and incompatible with gastroscopy. Conventional stimulated Raman scattering (SRS) microscopy has shown promise in label-free diagnosis on human tissues, yet it requires the tuning of picosecond lasers to achieve chemical specificity at the cost of time and complexity. Here, we demonstrate that single-shot femtosecond SRS (femto-SRS) reaches the maximum speed and sensitivity with preserved chemical resolution by integrating with U-Net. Fresh gastroscopic biopsy is imaged in <60 s, revealing essential histoarchitectural hallmarks perfectly agreed with standard histopathology. Moreover, a diagnostic neural network (CNN) is constructed based on images from 279 patients that predicts gastric cancer with accuracy >96%. We further demonstrate semantic segmentation of intratumor heterogeneity and evaluation of resection margins of endoscopic submucosal dissection (ESD) tissues to simulate rapid and automated intraoperative diagnosis. Our method holds potential for synchronizing gastroscopy and histopathological diagnosis.

The neuroprotective effects of fisetin, a natural flavonoid in neurodegenerative diseases: Focus on the role of oxidative stress

“…The immune reactive feature of Lewy bodies promotes mitochondrial damage that leads to progressive degeneration of nigrostriatal neurons. Dopamine secreting neurons of SN pars compacta exhibit a higher density of mitochondria that indicates the elevated rates of oxidative phosphorylation to meet the energy requirement ( Pacelli et al, 2015 ; Liu Z. et al, 2022 ). The SNc neurons increase ATP production up to threefold and also promote ROS generation ( Pacelli et al, 2015 ).…”

Intradialytic hypotension prediction using covariance matrix-driven whale optimizer with orthogonal structure-assisted extreme learning machine

“…Artificial intelligence (AI), which focuses on modeling human cognition in computing, has achieved significant progress in a broad range of disciplines ( Zhang J. et al, 2021 ; Luo et al, 2022 ). AI-assisted medical systems have recently gotten attention, making diagnosis systems and medical decision-making more instant, autonomous, and intelligent ( Li et al, 2020c , 2022 ; Zhang M. et al, 2021 ; Liu et al, 2022e ). Thus, developing an intelligent early-warning system to predict IDH will greatly assist HD staff in setting optimal dialysate and ultrafiltration parameters ( Lin et al, 2019 ).…”

DUnet: A deep learning guided protein-ligand binding pocket prediction

“…Convolutional neural network (CNN) is an effective and robust model for resolving segmentation problems [11]. Recently, CNN has been implemented for classification purposes in biological fields, such as for segmenting 2D bacterial images [12], diagnosis of gastroscopic biopsy [13] or (which was not certified by peer review) is the author/funder. All rights reserved.…”

“…Convolutional neural network (CNN) is an effective and robust model for resolving segmentation problems [11]. Recently, CNN has been implemented for classification purposes in biological fields, such as for segmenting 2D bacterial images [12], diagnosis of gastroscopic biopsy [13] or 3-dimensional (3D) MRI images [14]. For segmenting protein and protein-ligand binding sites in a 3D image, the reported 3D-CNNs models including Kalasanty, Deeppocket, and PUResNet showed better performance than the traditional physics-based methods [15][16][17].…”