Skip to content

Topic

Medical imaging AI

Machine learning applied to clinical images, covering acquisition, reconstruction, enhancement and interpretation in radiology and image-guided procedures.

Current stories

scienceOne report1 publisher

AMIFound, one restoration model for seven imaging modalities, improves 44 disease-detection tasks

AMIFound, a single model, outperformed 38 baselines at restoring degraded medical images across seven modalities, according to its authors. The downstream results are what make a shared restoration step ahead of clinical AI plausible.

Publishers:nature.com

Reality

Evidence58
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence55
buildOne report1 publisher

NVIDIA's 3D CT model leaves clinical validation to whoever post-trains it

NV-Reason-CT reads a 300-to-600-slice study as a single 3D volume instead of a stack of frames. NVIDIA published it as an uncleared research foundation whose cited multireader study measured its chest X-ray predecessor.

Reality

Evidence38
Adoption8
Hype gap+20
Incentives80
Confidence55
leadershipOne report1 publisher

Clinicians answer the AI cure-for-cancer claim with more than 100 brain tumour types

A Melbourne oncologist told the Guardian she doubts a cure for all cancer inside 10 years, and the clinicians who said so expect the real gains in diagnostics, pathology and treatment selection. Those are steps a budget can be written against.

Publishers:theguardian.com

Reality

Evidence62
Adoption30
Hype gap+58
Incentives55
Confidence64
buildOne report1 publisher

RADAR learns 146 abdominal CT findings from the reports radiologists already wrote

Alibaba's DAMO Academy put RADAR's training code and pretrained checkpoints under Apache 2.0 on September 18, a day after the Science paper. The 146 in the coverage counts radiological findings, cancers among them.

Publishers:runtimewire.com

Reality

Evidence70
Adoption20
Hype gap+20
Incentives70
Confidence62
buildOne report1 publisher

Comparing the crosshair with the sentence catches the left/right swaps in eight MRI studies

A developer ran eight signed radiology studies past Claude, Gemini, Grok and MedGemma, and the sides came back wrong often enough that he moved the laterality check out of the model and into the NIfTI affine.

Publishers:dev.to

Reality

Evidence45
Adoption10
Hype gap+12
Incentives55
Confidence55
scienceOne report1 publisher

A five-minute per-patient fit puts MIT's X-ray-to-CT registration under a millimeter

Existing 2D/3D registration models align well for some patients and fail for others. MIT's xvr, out today in Nature, fits itself to one patient in about five minutes, then matches that patient's X-rays to their 3D scan in seconds.

Publishers:news.mit.edu

Reality

Evidence58
Adoption
Not scored
Hype gap+30
Incentives70
Confidence55
investOne report1 publisher

Kakao Ventures seeds a four-month-old startup that estimates the angiography frames the X-ray skipped

Corologic's AngioFilm interpolates low-frame-rate coronary angiography video in real time so the machine takes fewer X-ray pictures, and the seed money from Kakao Ventures and SNU Holdings is going to multicenter validation and approvals.

Publishers:en.sedaily.com

Reality

Evidence30
Adoption20
Hype gap+20
Incentives70
Confidence50