Benchmark Pathology Report Text Corpus with Cancer Type Classification

Kefeli, Jenna; Tatonetti, Nicholas P.

doi:10.2139/ssrn.4418621

Cited by 1 publication

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we selected 9,523 pathology reports from The Cancer Genome Atlas (TCGA) [9,12]. The availability of TNM annotation in the TCGA metadata varied: 6,887 reports were documented with known tumor size (T), 5,678 reports with known regional lymph node involvement (N), and 4,608 reports with known metastasis (M).…”

Section: Resultsmentioning

confidence: 99%

“…Reports were initially stored in PDF format; in previous work, we converted the TCGA pathology report corpus to machine-readable plain text using OCR, performed extensive curation, and fully characterized the final TCGA report set. The final dataset spanned 9,523 reports, with 1:1 patient:report ratio [9,18]. TNM staging annotation was contained within the clinical metadata provided by TCGA [17].…”

Section: Tcga Pathology Report Dataset Construction With Tnm Annotationmentioning

confidence: 99%

“…Here, we utilize recent advances in natural language processing to classify cancer stage directly from pathology report text [9]. We specifically use a new variant of BERT [10][11], which has a larger input capacity than previous versions, and show that our model performs better than standard BERT models.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Generalizable and Automated Classification of TNM Stage from Pathology Reports with External Validation

Kefeli

Tatonetti

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Cancer staging is an essential clinical attribute informing patient prognosis and clinical trial eligibility. However, it is not routinely recorded in structured electronic health records. Here, we present a generalizable method for the automated classification of TNM stage directly from pathology report text. We train a BERT-based model using publicly available pathology reports across approximately 7,000 patients and 23 cancer types. We explore the use of different model types, with differing input sizes, parameters, and model architectures. Our final model goes beyond term-extraction, inferring TNM stage from context when it is not included in the report text explicitly. As external validation, we test our model on almost 8,000 pathology reports from Columbia University Medical Center, finding that our trained model achieved an AU-ROC of 0.815-0.942. This suggests that our model can be applied broadly to other institutions without additional institution-specific fine-tuning.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Tcga Pathology Report Dataset Construction With Tnm Annotationmentioning

confidence: 99%