High-speed photoacoustic and ultrasonic computed tomography of the breast tumor for early diagnosis with enhanced accuracy

We have developed a high-speed dual-modal imaging system (HDMI), designed to concurrently reveal anatomical and hematogenous details of the human breast within seconds. Through innovative system design and technical advancements, HDMI integrates large-view photoacoustic and ultrasonic computed tomography with standardized scanning and batch data processing for computer-aided diagnosis. It achieves dual-modal imaging at a 10 Hz frame rate and completes a whole-breast scan in 12 seconds, providing penetration up to 5 cm in vivo. In a clinical study involving 170 patients with 186 breast tumors, we developed a diagnostic model leveraging combined photoacoustic and ultrasound features. In a triple-blinded comparison using pathological diagnosis as the ground truth, HDMI significantly improved diagnostic specificity from 22.5% to 75.0% compared to clinical ultrasonography. This technology shows strong potential for early breast tumor diagnosis, offering enhanced accuracy without the need for..., This study was approved by the Ethics Committees of Zhejiang University (2023-001) and the First Affiliated Hospital, Zhejiang University School of Medicine (IIT20230241B). All participants signed informed consent forms. All HDMI data files were de-identified, with personal identifiers removed and coded labels used to ensure confidentiality., , # High-speed photoacoustic and ultrasonic computed tomography of the breast tumor for early diagnosis with enhanced accuracy

Dataset DOI: [10.5061/dryad.vmcvdnd55](10.5061/dryad.vmcvdnd55)

## Description of the data and file structure

We shared Photoacoustic Computed Tomography (PACT) and Ultra-Resolution Computed Tomography (URCT) imaging data of 161 human breast tumors (ground truth population), capturing both anatomical and hematogenous characteristics of the lesions.

### Files and variables

#### File: DATA.zip

**Description:**  The PACT and URCT images of 161 human breast tumors, comprising a total of 322 “.mat” files.

## Code/software

All data files can be opened in MATLAB.