Search

Abstract Malignancy grading of prostate cancer (PCa) is fundamental for risk stratification, patient counseling, and treatment decision-making. Deep learning has shown potential to improve the expert consensus for tumor grading, which relies on the Gleason score/grade grouping. However, the core problem of interobserver variability for the Gleason grading system remains unresolved. We developed a novel grading system for PCa and utilized artificial intelligence (AI) and multi-institutional international datasets from 2647 PCa patients treated with radical prostatectomy with a long follow-up of ≥10 years for biochemical recurrence and cancer-specific death. Through survival analyses, we evaluated the novel grading system and showed that AI could develop a tumor grading system with four risk groups independent from and superior to the current five grade groups. Moreover, AI could develop a scoring system that reflects the risk of castration resistant PCa in men who have experienced biochemical recurrence. Thus, AI has the potential to develop an effective grading system for PCa interpretable by human experts.

Purpose: Entering all information from narrative documentation for clinical research into databases is time consuming, costly, and nearly impossible. Even high-volume databases do not cover all patient characteristics and drawn results may be limited. A new viable automated solution is machine learning based on deep neural networks applied to natural language processing (NLP), extracting detailed information from narratively written (eg, pathologic radical prostatectomy [RP]) electronic health records (EHRs)., Methods: Within an RP pathologic database, 3,679 RP EHRs were randomly split into 70% training and 30% test data sets. Training EHRs were automatically annotated, providing a semiautomatically annotated corpus of narratively written pathologic reports with initially context-free gold standard encodings. Primary and secondary Gleason pattern, corresponding percentages, tumor stage, nodal stage, total volume, tumor volume and diameter, and surgical margin were variables of interest. Second, state-of-the-art NLP techniques were used to train an industry-standard language model for pathologic EHRs by transfer learning. Finally, accuracy of the named entity extractors was compared with the gold standard encodings., Results: Agreement rates (95% confidence interval) for primary and secondary Gleason patterns each were 91.3% (89.4 to 93.0), corresponding to the following: Gleason percentages, 70.5% (67.6 to 73.3) and 80.9% (78.4 to 83.3); tumor stage, 99.3% (98.6 to 99.7); nodal stage, 98.7% (97.8 to 99.3); total volume, 98.3% (97.3 to 99.0); tumor volume, 93.3% (91.6 to 94.8); maximum diameter, 96.3% (94.9 to 97.3); and surgical margin, 98.7% (97.8 to 99.3). Cumulative agreement was 91.3%., Conclusion: Our proposed NLP pipeline offers new abilities for precise and efficient data management from narrative documentation for clinical research. The scalable approach potentially allows the NLP pipeline to be generalized to other genitourinary EHRs, tumor entities, and other medical disciplines.