Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
Letter to the Editor: Potential pitfalls in deep learning-based imaging for spontaneous intracerebral hemorrhage
Sachin D Balutkar, Department of Radiology, Maharashtra Institute of Medical Sciences and Research (Medical College), Latur 413512, Maharashtra, India
Sachin S Bhavthankar, Department of Biochemistry, Maharashtra Institute of Medical Sciences and Research, Latur 413512, Maharashtra, India
Basavraj S Nagoba, Department of Microbiology, Maharashtra Institute of Medical Sciences and Research (Medical College), Latur 413531, Maharashtra, India
Author contributions: Balutkar SD and Bhavthankar SS drafted the primary manuscript; Nagoba BS provided conceptual guidance, performed critical revisions, and edited the text; All the authors participated in the final drafting and have reviewed and approved the submitted version.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Corresponding author: Basavraj S Nagoba, PhD, Professor, Department of Microbiology, Maharashtra Institute of Medical Sciences and Research (Medical College), Vishwanathpuram, Ambajogai Road, Latur 413531, Maharashtra, India. basavraj.nagoba@mimsr.edu.in
Received: February 24, 2026
Revised: March 9, 2026
Accepted: May 18, 2026
Published online: July 28, 2026
Processing time: 160 Days and 1.8 Hours
Revised: March 9, 2026
Accepted: May 18, 2026
Published online: July 28, 2026
Processing time: 160 Days and 1.8 Hours
Core Tip
Core Tip: This critique evaluates a study using radiomics and deep learning to predict outcomes in spontaneous intracerebral hemorrhage. While the original study shows promise, it may be limited by confounding calcifications and chronic ischemic lesions being misidentified as hematoma or edema due to broad Hounsfield units thresholds. Furthermore, the use of limited axial slices rather than full volumetric analysis may reduce accuracy. We propose an integrated approach that combines deep learning efficiency with strict Hounsfield units filtering and expert radiological oversight to minimize false positives and improve predictive precision.