Localization and classification of intracranialhemorrhages in CT data

TL;DR

This paper presents an automatic CNN-based system for detecting and classifying intracranial hemorrhages in CT scans, aiming to reduce diagnostic time in acute cases.

Contribution

It introduces a novel cascade-parallel CNN architecture for ICH detection and classification, demonstrating promising accuracy on a public dataset.

Findings

Average Jaccard coefficient of 53.7% on CQ500 dataset

Potential to significantly decrease diagnostic duration

Effective localization and classification of ICHs

Abstract

Intracranial hemorrhages (ICHs) are life-threatening brain injures with a relatively high incidence. In this paper, the automatic algorithm for the detection and classification of ICHs, including localization, is present. The set of binary convolutional neural network-based classifiers with a designed cascade-parallel architecture is used. This automatic system may lead to a distinct decrease in the diagnostic process's duration in acute cases. An average Jaccard coefficient of 53.7 % is achieved on the data from the publicly available head CT dataset CQ500.