Skip to content
Partaw Research
Menu

The evidence gap is the invitation

Teleradiology case-finding in rural Afghanistan has, as far as we have found, never been prospectively described. The platform generating the data is running. What the study needs now is its partners.

Three concrete offers


Co-authorship

A prospective study of teleradiology case-finding in rural Afghanistan — a setting with, as far as we have found, no prospective description in the literature. For global-health and radiology faculty who need LMIC scholarship that is genuinely novel rather than incremental, the data source is running and the authorship is open.


Founding advisory seats

The advisory board is forming now. Seats cover clinical governance, research design, and data governance — named roles with real influence over protocols, escalation rules, and what the platform refuses to do. Founding advisors shape the program rather than endorse it.


Dataset access & a validation site

Governed access to a growing, de-identified corpus of rural Afghan radiology — including smartphone photographs of physical films, the image reality of most of the world — and a real-world deployment site for validating imaging software on a population its training data has never seen.

Three kinds of partner

Academic radiology & global-health programs
University radiology departments and global-health imaging programs: adjudication partnership, resident and fellow involvement, co-authorship, advisory seats.
CAD & imaging-AI teams
Validation partnerships on out-of-distribution, real-world data — with results published, not marketed.
TB-ecosystem NGOs & funders
Implementers and funders working on case-finding in high-burden, low-resource settings: a deployment partner whose reading pathway is already running.

The deepest current opportunity is the TB screening program — its architecture, limits, and open questions have their own page, written to be sent on its own.

The dataset is a public-good asset. It is governed like one.

Consent is collected at intake, before a film enters any research use.

De-identification happens before research use — identifying details are stripped at the platform boundary, and patient data never appears on this site.

Dataset access is granted under data-governance agreements, for legitimate research purposes. The corpus is a public-good asset: it is never sold.

Every study protocol is designed for ethics review with our academic partners’ institutions — we bring the setting and the data pipeline; partners bring the review pathway their institutions require.

The study runs on infrastructure operated by Partaw’s commercial arm, disclosed in every protocol and publication and kept at arm’s length.

A short note is enough to start

Protocols and governance drafts follow in conversation. A human reads every message.