Lightweight deep learning framework for intracranial hemorrhage detection in brain CT scans
Sumaira Hussain, Salman Jan, Manal Aldhayan, Mohammad Asmat Ullah Khan, Shahid Kamal, Abid Jameel, Jawad Hasan Alkhateeb, Jamal Zraqou · 2026
Read Lightweight deep learning framework for intracranial hemorrhage detection in brain CT scans by Sumaira Hussain, Salman Jan, Manal Aldhayan, Mohammad Asmat Ullah Khan, Shahid Kamal, Abid Jameel, Jawad Hasan Alkhateeb, Jamal Zraqou (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13318755.txt.
Description
Sumaira Hussain and co-authors present a 2026 study of lightweight deep learning for spotting brain bleeds on CT scans. Intracranial haemorrhage is life-threatening and needs fast reading. Manual review is slow and varies between readers. Small models could help in clinics with limited computing power.
The team compared a custom small network with two mobile designs, MobileNetV2 and EfficientNet-B0. Data came from a public brain CT haemorrhage dataset. MobileNetV2 performed best, reaching 87 percent accuracy and an AUC of 0.94 in their tests. The custom network reached 79 percent accuracy. EfficientNet-B0 competed well but needed more computing resources. The authors back such models as decision support, pending better sensitivity and clinical validation.
Catalog record
| CREATOR | Sumaira Hussain, Salman Jan, Manal Aldhayan, Mohammad Asmat Ullah Khan, Shahid Kamal, Abid Jameel, Jawad Hasan Alkhateeb, Jamal Zraqou |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35640 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |