MSD and LaNova Medicines Enter Exclusive Global License Agreement for Novel Cancer Therapy, LM-299

Promising CAR T Cell Therapies from BMS and Cabaletta Bio Show Potential for Durable Remission in Severe Lupus

FDA Grants Accelerated Approval to PTC Therapeutics' KEBILIDI™, the First Gene Therapy for AADC Deficiency in the U.S.

Association for Molecular Pathology Publishes Joint Consensus Report on SARS-CoV-2 Genomic Surveillance for Hospitals and Public Health Laboratories

Philips India Unveils Advanced Azurion Imaging System to Revolutionize Cardiac and Neurovascular Care

Tempus AI Study Highlights Benefits of Dual RNA and DNA Sequencing for Advanced Lung Cancer

Ontada and Datavant Partner to Expand Access to Real-World Oncology Data for Accelerated Cancer Innovation

Tempus AI and Flatiron Health Partner to Integrate Genomic Testing into OncoEMR for Enhanced Oncology Care

Flare Therapeutics and Roche Forge Strategic Partnership to Discover Precision Oncology Therapies Targeting Transcription Factors

FDA Approves Autolus’s CAR-T Therapy Aucatzyl for Relapsed or Refractory B-ALL in Adults

Biosimilars: Expanding Access to Precision Medicine through Affordable Biologic Therapies

IDEAYA Biosciences Secures Exclusive Worldwide License for Biocytogen’s BsADC Program IDE034 for Solid Tumor Treatment

Seer and Thermo Fisher Scientific Announce Strategic Collaboration to Advance Proteomics with Integrated Platform Solution

SystImmune Secures FDA Clearance for IND Application of Novel ADC Therapy BL-M17D1 for Advanced Solid Tumors

Medicare Expands Coverage for Adaptive Biotechnologies' clonoSEQ® MRD Test to Include Mantle Cell Lymphoma

Home
/
update
/
details

Researchers Validate Deep Bio's AI Model for Prostate Cancer Assessment

NextEdge Admin

07 Aug 2024

2 min 36 sec

Deep Bio Inc., a leader in AI-driven digital pathology, announced the publication of an external validation study for its DeepDx Prostate AI algorithm, conducted by Stanford Medicine. Published in the British Journal of Urology International, the study titled “External Validation of an Artificial Intelligence Model for Gleason Grading of Prostate Cancer on Prostatectomy Specimens” evaluated the algorithm's performance on whole-mount radical prostatectomy (RP) specimens.

The DeepDx Prostate algorithm, originally trained on prostate core needle biopsy (CNB) images from South Korean hospitals, was tested on RP specimens from a different institution without any fine-tuning. Key findings include:

1. High Accuracy: The algorithm achieved a sensitivity of 0.997 and specificity of 0.88 in detecting cancer in RP specimens.
2. Agreement with Pathologists: It showed strong agreement with expert uropathologists, with Cohen’s Kappa values of 0.91 for cancer presence, 0.89 for Gleason grade classification, and 0.89 for risk group identification.
3. Generalizability: The algorithm demonstrated robust performance across different datasets and tissue types, suggesting its potential for widespread implementation.

Sun Woo Kim, CEO of Deep Bio, emphasized the importance of accurate Gleason grading for optimizing treatment plans. Bogdana Schmidt MD, MPH, the lead author and assistant professor at the University of Utah, highlighted the algorithm's generalizability and potential for clinical use, noting its value in timely and accurate grading of prostate cancer.

The study concludes that DeepDx Prostate is a highly accurate tool for identifying and grading prostate cancer on digital histopathology images, showing almost perfect concordance with expert pathologists. DeepDx Prostate also demonstrates 99% sensitivity and 97% specificity when used on CNB samples.