Acute Appendicitis

Value Proposition

Acute appendicitis is the most common abdominal emergency and often requires surgical intervention. Most cases, if positive on CT, are heading to the OR, though some patients are treated with antibiotics for early appendicitis. The rapid identification of acute appendicitis can streamline these processes and ensure people get the treatment they need. An algorithm meeting this use case would also help prioritize the case on the worklist.

Narrative(s)

A 25-year-old male presents to the emergency room with fever and abdominal pain. The patient has been getting progressively worse over the past few days. Associated lab findings are leukocytosis (elevated white blood cell count) and anorexia.

Workflow Description

An algorithm may be written to query patient records and identify eating disorders, lab tests, or other patient records that may indicate a patient is susceptible to acute appendicitis and aid in the diagnosis.

An image is obtained from the modality and sent to PACS and the AI engine. The image is analyzed by the engine. The system categorizes the status of the appendix as normal, ruptured appendicitis, nonruptured appendicitis, or undetermined. A message is sent to PACS from the engine with the classification information. Ancillary data is also sent to the PACS if collected by the engine.

Procedures(s): {CT, CECT}

Sex at Birth: {Male, Female}

Age: [0,90]

Obstruction: {Stone, fecalith, tumor, other mass}

Peri-appendiceal: {Fat stranding, free fluid, abscess}

Lumen Contents: {Air-filled, contrast-filled, fluid-filled}

Appendicitis: {Ruptured, nonruptured}

Inputs

DICOM Study

Primary Outputs

Appendicitis Detection

Secondary Outputs

Appendiceal Diameter

Peri-appendiceal Fat Stranding

Free Fluid

Obstructing Focus

Lumen Contents

Peri-appendiceal Abscess

Opacification

The Cancer Imaging Archive (TCIA)

Know of more related datasets? Please let us know