Abstract
BackgroundHelicobacter pylori, a 2 × 1 μm spiral-shaped bacterium, is the most common risk factor for gastric cancer worldwide. Clinically, patients presenting with symptoms of gastritis, routinely undergo gastric biopsies. The following histo-morphological evaluation dictates therapeutic decisions, where antibiotics are used for H. pylori eradication. There is a strong rational to accelerate the detection process of H. pylori on histological specimens, using novel technologies, such as deep learning.MethodsWe designed a deep-learning-based decision support algorithm that can be applied on regular whole slide images of gastric biopsies. In detail, we can detect H. pylori both on Giemsa- and regular H&E stained whole slide images.ResultsWith the help of our decision support algorithm, we show an increased sensitivity in a subset of 87 cases that underwent additional PCR- and immunohistochemical testing to define a sensitive ground truth of HP presence. For Giemsa stained sections, the decision support algorithm achieved a sensitivity of 100% compared to 68.4% (microscopic diagnosis), with a tolerable specificity of 66.2% for the decision support algorithm compared to 92.6 (microscopic diagnosis).ConclusionTogether, we provide the first evidence of a decision support algorithm proving as a sensitive screening option for H. pylori that can potentially aid pathologists to accurately diagnose H. pylori presence on gastric biopsies.
Highlights
Helicobacter pylori, a 2 × 1 μm spiral-shaped bacterium, is the most common risk factor for gastric cancer worldwide
We aimed to (1) design a deep learning based decision support algorithm that highlights H. pylori bacteria in image regions of gastric biopsies samples that are routinely tested for H. pylori presence and (2) validate this algorithm both on Giemsa stains and regular H&E stains comparing with microscopic diagnosis, immunohistochemistry and Polymerase chain reaction (PCR)
Detection of H. pylori: image processing Our approach consists of first localizing areas of H. pylori presence using image processing, and cropping the hot spots into 224 × 224 patches and classifying them with a deep neural network
Summary
Helicobacter pylori, a 2 × 1 μm spiral-shaped bacterium, is the most common risk factor for gastric cancer worldwide. Advances in the field of hardware components, as well as the availability of large amounts of data, have allowed the field of artificial intelligence to rapidly grow [17] These technologies have been successfully applied to improve diagnostic procedures in the medical field [1,2,3]. We aimed to (1) design a deep learning based decision support algorithm that highlights H. pylori bacteria in image regions of gastric biopsies samples that are routinely tested for H. pylori presence and (2) validate this algorithm both on Giemsa stains and regular H&E stains comparing with microscopic diagnosis, immunohistochemistry and PCR
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
