DSAI-09 INTERPRETABLE RADIOMICS RISK SIGNATURE FOR DISTINGUISHING TUMOR RECURRENCE FROM RADIATION NECROSIS IN BRAIN METASTASES ON POST-TREATMENT MRI: A COMPARATIVE ANALYSIS WITH ‘BLACK-BOX’ APPROACHES
Dheerendranath Battalapalli, Hyemin Um, Marwa Ismail, Virginia B Hill, Sushant Puri, Jennifer S Yu, Lan Lu, Ameya P Nayate, Anthony Higinbotham, Lisa R Rogers, Prateek Prasanna, Mainak Bardhan, Chengnan Li, Mustafa M Basree, Andrew M Baschnagel, Alan B McMillan, Ankush Bhatia, Manmeet S Ahluwalia, Michael C Veronesi, Pallavi Tiwari · 2026
Read DSAI-09 INTERPRETABLE RADIOMICS RISK SIGNATURE FOR DISTINGUISHING TUMOR RECURRENCE FROM RADIATION NECROSIS IN BRAIN METASTASES ON POST-TREATMENT MRI: A COMPARATIVE ANALYSIS WITH ‘BLACK-BOX’ APPROACHES by Dheerendranath Battalapalli, Hyemin Um, Marwa Ismail, Virginia B Hill, Sushant Puri, Jennifer S Yu, Lan Lu, Ameya P Nayate, Anthony Higinbotham, Lisa R Rogers, Prateek Prasanna, Mainak Bardhan, Chengnan Li, Mustafa M Basree, Andrew M Baschnagel, Alan B McMillan, Ankush Bhatia, Manmeet S Ahluwalia, Michael C Veronesi, Pallavi Tiwari (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/PMC13448279.txt.
Description
Battalapalli and colleagues present in 2026, as abstract DSAI-09, a head-to-head test of an interpretable imaging signature against two black-box artificial intelligence models. The puzzle is classic: telling recurrent tumor from radiation necrosis, dead tissue from earlier radiosurgery, on MRI scans of treated brain metastases.
The team assembled 233 scans from three institutions, most with pathology confirmation, and computed 944 texture and spatial features per lesion for their SMART-Risk signature. On the held-out Wisconsin test set it reached an accuracy in the high seventies. It clearly beat a standard deep network and a fine-tuned medical imaging foundation model, which managed only the low-to-mid sixties. Importance analysis showed recurrent tumors display greater spatial complexity than necrotic tissue, giving clinicians an explanation, not a bare score. The authors argue such transparency could cut unnecessary surgery for what is only treatment damage.
Catalog record
| CREATOR | Dheerendranath Battalapalli, Hyemin Um, Marwa Ismail, Virginia B Hill, Sushant Puri, Jennifer S Yu, Lan Lu, Ameya P Nayate, Anthony Higinbotham, Lisa R Rogers, Prateek Prasanna, Mainak Bardhan, Chengnan Li, Mustafa M Basree, Andrew M Baschnagel, Alan B McMillan, Ankush Bhatia, Manmeet S Ahluwalia, Michael C Veronesi, Pallavi Tiwari |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 3 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |