QP-Prostate® CAD

QP-Prostate® CAD is artificial intelligence-based software that automatically detects and identifies suspected lesions in the prostate gland from bi-parametric MRI (T2-weighted and DWI sequences). It is intended as a concurrent read aid for trained physicians interpreting prostate MRI studies, displaying results as colored overlay markers indicating moderate or high suspicion levels for clinically significant prostate cancer. Patient management decisions should not be based solely on the device output.

Both QP-Prostate® CAD and the predicate device (ProstatID) are DICOM-compatible CADe/CADx software using T2-weighted and DWI sequences as input. Key differences: QP-Prostate uses Neural Networks and Machine Learning algorithms whereas ProstatID uses Random Forest; QP-Prostate provides two suspicion levels (moderate/high) via colored markers while ProstatID provides continuous 0-1 scale scores; QP-Prostate does not include ADC data in analysis while ProstatID does. Both generate 3D renditions with DICOM output to PACS workstations.

The device was developed according to FDA recognized consensus standards for software development. Software verification and validation was performed following V&V plans and protocols verifying that product specifications were met. Not stated in this summary are specific ISO, IEC, or ASTM standard citations.

QP-Prostate® CAD and ProstatID have substantially similar intended uses (concurrent reading aid for prostate MRI interpretation), same regulatory classification (Class II, 21 CFR 892.2090, product code QDQ), identical input methodology (DICOM T2W/DWI), and equivalent patient populations (patients 40+ years undergoing prostate MRI) and user qualifications. Although algorithmic approaches differ (Neural Networks vs. Random Forest) and output formats vary (two suspicion levels vs. continuous scale), performance testing demonstrated QP-Prostate® CAD achieves statistically significant clinical reader performance improvement (p=0.039, ΔAUC=0.019) comparable to ProstatID. These technological differences do not raise different safety or effectiveness questions because both devices undergo independent verification and validation demonstrating equivalent diagnostic performance.