Is Mamba Reliable for Medical Imaging?
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910026532126720 |
|---|---|
| author | Latibari, Banafsheh Saber Nazari, Najmeh Brignac, Daniel Sayadi, Hossein Homayoun, Houman Mahalanobis, Abhijit |
| author_facet | Latibari, Banafsheh Saber Nazari, Najmeh Brignac, Daniel Sayadi, Hossein Homayoun, Houman Mahalanobis, Abhijit |
| contents | State-space models like Mamba offer linear-time sequence processing and low memory, making them attractive for medical imaging. However, their robustness under realistic software and hardware threat models remains underexplored. This paper evaluates Mamba on multiple MedM-NIST classification benchmarks under input-level attacks, including white-box adversarial perturbations (FGSM/PGD), occlusion-based PatchDrop, and common acquisition corruptions (Gaussian noise and defocus blur) as well as hardware-inspired fault attacks emulated in software via targeted and random bit-flip injections into weights and activations. We profile vulnerabilities and quantify impacts on accuracy indicating that defenses are needed for deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_16723 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Is Mamba Reliable for Medical Imaging? Latibari, Banafsheh Saber Nazari, Najmeh Brignac, Daniel Sayadi, Hossein Homayoun, Houman Mahalanobis, Abhijit Cryptography and Security Artificial Intelligence State-space models like Mamba offer linear-time sequence processing and low memory, making them attractive for medical imaging. However, their robustness under realistic software and hardware threat models remains underexplored. This paper evaluates Mamba on multiple MedM-NIST classification benchmarks under input-level attacks, including white-box adversarial perturbations (FGSM/PGD), occlusion-based PatchDrop, and common acquisition corruptions (Gaussian noise and defocus blur) as well as hardware-inspired fault attacks emulated in software via targeted and random bit-flip injections into weights and activations. We profile vulnerabilities and quantify impacts on accuracy indicating that defenses are needed for deployment. |
| title | Is Mamba Reliable for Medical Imaging? |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2602.16723 |