Aidoc has received U.S. FDA Breakthrough Device Designation for First Read, an investigational AI tool designed to analyse chest radiographs and generate preliminary radiology report text.

The designation applies to a broad chest radiograph analysis solution for four life-threatening findings. First Read has not yet been cleared or approved by the FDA and remains for investigational use.

This is relevant because radiology departments are facing rising imaging volumes, workforce shortages and growing pressure to reduce turnaround times. Reporting remains a major bottleneck, particularly when demand increases faster than radiologist capacity.

Aidoc said First Read is designed to generate high-quality draft text from imaging findings. The aim is to reduce time spent on reporting tasks and return more capacity to radiologists for interpretation, judgement and patient care.

The tool builds on Aidoc’s wider clinical AI platform, which supports diagnostic decisions across nearly 2,000 hospitals worldwide. The company said its platform has analysed more than 120 million patient cases, giving the technology a large clinical deployment base.

First Read is built on the same underlying architecture as Aidoc’s FDA-cleared abdominal CT triage application. This is important because generative AI in radiology must be supported by safety, quality and reliability frameworks, especially when output becomes part of clinical documentation.

The clinical workflow angle is central. Radiology AI has historically focused on detection, triage or prioritisation. First Read moves further downstream into reporting, where errors, missing details and automation bias can have clinical, operational and legal consequences.

The FDA Breakthrough Device pathway is intended to accelerate development and review of technologies that may provide more effective diagnosis or treatment for serious conditions and address unmet clinical needs. For Aidoc, the designation may support closer regulatory interaction as it develops First Read.

Hospitals will also need to consider governance around generative AI outputs. Draft report systems must be evaluated for accuracy, consistency, explainability, auditability and performance across patient populations, imaging equipment and institutional reporting styles.

The development reflects how AI in medical imaging is moving from point-solution detection toward broader workflow assistance. If clinically validated and cleared, AI-generated draft reporting could become part of a wider effort to make radiology operations more sustainable as imaging demand continues to rise.