AI for Ovarian Cancer Detection
Precision Care Through AI
Founded in 2019 as a spin-off from the Polytechnic University of Turin, SynDiag develops AI-powered solutions for advanced sonographic diagnostics. Our flagship product, OvAi, is a certified Software as a Medical Device that supports ovarian cancer assessment through two modules: OvAi Focus for anatomical analysis and OvAi X for Virtual Biopsy. We're now expanding our technology to address uterine cancer and beyond, driving the future of predictive and objective sonography.
Diagnostic Challenges
Ultrasound is the primary tool for diagnosing ovarian cancer, as biopsies are not feasible. However, the accuracy of ultrasound is highly dependent on the operator’s expertise. This variability contributes to late diagnoses in 75% of cases—leading to survival rates below 40%. In contrast, early detection can raise survival chances to over 80%, highlighting the urgent need for more objective and reliable diagnostic tools.
🟥 Late Diagnosis
75% of cases - 40% survival rate
0
%
🟩 Early Diagnosis
Early stage detection
0
%
Solution
OvAi is a cloud-based web application that enables early detection of ovarian cancer through ultrasound imaging. Compatible with any ultrasound machine or PACS system, OvAi analyzes ultrasound videos and delivers accurate, AI-driven diagnostic insights in just 5 minutes—supporting a fast, reliable, and all-in-one diagnostic workflow.
AI for Anatomical
Description, is designed
for standard reporting
AI for Virtual Biopsy
able to screen which
kind of tumor is under
investigation
Management and
Telemedicine system
for a second opinion
request in real time
An archive of
diagnosed clinical
cases available for free
consultation
Advanced AI for Ovarian Cancer Imaging
OvAi Focus
Morphology Analysis: Identifies lesion contours from B-Mode and Doppler videos with a DICE coefficient of 88%, ensuring high accuracy in shape recognition.
Vascularization Enhancement: Suppresses Doppler artifacts in ultrasound videos, resulting in a 24% increase in video quality for clearer vascular imaging.
OvAi X
Virtual Biopsy Performance: Achieves an AUC (Area Under Curve) of 94.8%, indicating exceptional diagnostic accuracy for non-invasive cancer characterization.
SynDiag
Partners
Predictable histologies
OvAi is designed to predict a wide range of ovarian tumor histologies with high accuracy. These include benign conditions such as serous cystadenoma, mucinous cystadenoma, cystadenofibroma, the thecoma-fibroma group, endometrioma, and dermoid cysts. It also identifies borderline tumors like serous borderline and mucinous borderline tumors. Additionally, OvAi supports the detection of malignant histologies, including high-grade serous adenocarcinoma, low-grade serous adenocarcinoma, mucinous adenocarcinoma, endometrioid adenocarcinoma, clear cell carcinoma, and metastatic tumors.
