FDA clearedFDA cleared. Stage 4 of 4 (Authorized)

FDA Clears Ever Fortune.AI Software That Flags Suspected Appendicitis on CT

The Taiwanese company's triage software reported 92.8% sensitivity and 93.2% specificity in a 300-case U.S. validation study described in its 510(k) summary.

Volume-rendered contrast CT image of the abdomen and pelvis
A volume rendering of a contrast CT of the abdomen and pelvis showing normal anatomy. Not an image from the software described in this article. Mikael Häggström, M.D., CC0, via Wikimedia Commons · CC0
FDA cleared510(k) · K253163FDA cleared. Stage 4 of 4 (Authorized)

The FDA cleared this device through the 510(k) pathway, which finds it substantially equivalent to a device already on the market. Clearance is not the same as approval.

Key points

  • The FDA cleared EFAI ERSUITE CT Appendicitis Assessment System (K253163) on April 22 as Class II radiological computer-aided triage and notification software.
  • The software analyzes contrast-enhanced abdominal CT and sends text notifications for cases with suspected appendicitis; it is not intended as a diagnostic device.
  • The company's 510(k) summary reports 92.8% sensitivity and 93.2% specificity on 300 U.S. cases, against performance goals of 80%.

The U.S. Food and Drug Administration on April 22 cleared artificial intelligence software from Taiwan's Ever Fortune.AI that analyzes contrast-enhanced abdominal CT scans and alerts clinicians to cases with suspected appendicitis, according to the agency's 510(k) database entry and clearance letter.

The product, EFAI ERSUITE CT Appendicitis Assessment System (APPEN-CT-100), was found substantially equivalent under 510(k) number K253163. FDA records list it as a Class II device under regulation 21 CFR 892.2080, radiological computer-aided triage and notification software, product code QAS. The database shows the submission was received on September 26, 2025, and reviewed as a traditional 510(k), without a third-party reviewer.

What the software does

According to the indications for use, the software is intended to help hospital networks and trained clinicians with workflow triage by flagging and communicating suspected positive findings of appendicitis. It runs in parallel with standard image interpretation and presents notifications for cases with suspected findings.

The device "does not alter the original medical image and is not intended to be used as a diagnostic device," the indications state. Clinicians who receive a notification remain responsible for viewing the full images under the standard of care. The 510(k) summary says the software does not mark or highlight a specific location on the image, and does not remove or deprioritize cases from the reading queue. It sends text-based notifications to the PACS or radiology workstation.

Ever Fortune.AI named BriefCase (K193298), a triage tool for intra-abdominal free gas on abdominal CT, as its predicate device. The company said the two products differ in the finding they look for and in contrast use, and that its notifications contain no preview images, but that these differences do not raise new questions of safety or effectiveness.

Validation results

The company ran a retrospective, blinded, multisite study on 300 consecutively collected contrast-enhanced abdominal CT cases from several U.S. clinical sites, one case per patient, none of which had been used to develop or internally test the model, according to the 510(k) summary. Three U.S. board-certified radiologists assessed each case, and their majority view set the reference standard, giving 139 positive and 161 negative cases.

The software showed sensitivity of 0.928 (95% confidence interval 0.873 to 0.960) and specificity of 0.932 (0.882 to 0.961), above the pre-specified goal of 0.8 for each. Mean processing time per case was 3.37 minutes. The company reported consistent performance in subgroups by sex, age, race and ethnicity, CT manufacturer and slice thickness. The figures are as reported by the company in its 510(k) summary.

The FDA's clearance letter was signed by Jessica Lamb, assistant director of the Division of Radiological Imaging Devices and Electronic Products in the Center for Devices and Radiological Health.

Sources

  1. 510(k) Premarket Notification K253163: EFAI ERSUITE CT Appendicitis Assessment System (APPEN-CT-100), FDA, April 22, 2026
  2. K253163 clearance letter, indications for use and 510(k) summary, FDA, April 22, 2026