Comparative evaluation of convolutional and transformer-based object detection models for brain tumor localization in MRI
Farhan Naeem, Amara Haroon, Nasru Minallah, Tufail Ahmad, Shahi Dost · 2026
Read Comparative evaluation of convolutional and transformer-based object detection models for brain tumor localization in MRI by Farhan Naeem, Amara Haroon, Nasru Minallah, Tufail Ahmad, Shahi Dost (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/plos-pone-0358483.txt.
Description
Farhan Naeem and four colleagues pit two modern object detection systems against each other on brain scans, in a 2026 PLOS ONE study. Finding a tumor's exact location on magnetic resonance images is a step toward early diagnosis and treatment planning. The team compared a convolution based detector, YOLOv11s, with a transformer based rival, RF-DETR, on the same annotated scan collection.
YOLOv11s came out ahead on the headline measures, with detection quality, precision, and recall all a few points higher. The transformer model held one advantage: it located tumors more tightly when judged by stricter overlap standards. The authors note the hard part was fairness, since the two systems learn features differently and train under different settings. The comparison gives imaging teams a grounded basis for choosing between the two families.
Catalog record
| CREATOR | Farhan Naeem, Amara Haroon, Nasru Minallah, Tufail Ahmad, Shahi Dost |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33948 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |