Transforming Thyroid Cancer Diagnosis and Staging Information from Unstructured Reports to the Observational Medical Outcome Partnership Common Data Model

Abstract

Background Cancer staging information is an essential component of cancer research. However, the information is primarily stored as either a full or semistructured free-text clinical document which is limiting the data use. By transforming the cancer-specific data to the Observational Medical Outcome Partnership Common Data Model (OMOP CDM), the information can contribute to establish multicenter observational cancer studies. To the best of our knowledge, there have been no studies on OMOP CDM transformation and natural language processing (NLP) for thyroid cancer to date. Objective We aimed to demonstrate the applicability of the OMOP CDM oncology extension module for thyroid cancer diagnosis and cancer stage information by processing free-text medical reports. Methods Thyroid cancer diagnosis and stage-related modifiers were extracted with rule-based NLP from 63,795 thyroid cancer pathology reports and 56,239 Iodine whole-body scan reports from three medical institutions in the Observational Health Data Sciences and Informatics data network. The data were converted into the OMOP CDM v6.0 according to the OMOP CDM oncology extension module. The cancer staging group was derived and populated using the transformed CDM data. Results The extracted thyroid cancer data were completely converted into the OMOP CDM. The distributions of histopathological types of thyroid cancer were approximately 95.3 to 98.8% of papillary carcinoma, 0.9 to 3.7% of follicular carcinoma, 0.04 to 0.54% of adenocarcinoma, 0.17 to 0.81% of medullary carcinoma, and 0 to 0.3% of anaplastic carcinoma. Regarding cancer staging, stage-I thyroid cancer accounted for 55 to 64% of the cases, while stage III accounted for 24 to 26% of the cases. Stage-II and -IV thyroid cancers were detected at a low rate of 2 to 6%. Conclusion As a first study on OMOP CDM transformation and NLP for thyroid cancer, this study will help other institutions to standardize thyroid cancer–specific data for retrospective observational research and participate in multicenter studies.

A Systematic Review of Natural Language Processing Methods and Applications in Thyroidology

“…Three studies used both radiology and pathology reports. 12 , 27 , 35 The studies were presented with varying data set sizes. Of the 24 studies, 11 (45%) used sample sizes of less than 5000 records, and 9 of 24 (38%) studies used data sets with more than 10,000 records.…”

“…Lian et al 25 developed an NLP pipeline to measure and categorize health-related quality of life based on narrative interviews with patients who underwent surgical treatment (area under the curve, 0.76; accuracy, 70.09; SN, 70.02%; and PPV, 70.20%). 25 In addition, Yoo et al 12 developed an NLP algorithm to determine thyroid cancer diagnosis and stage based on retrospective information from medical records, specifically using surgical pathology and whole-body scan reports (SN, 100%; PPV, 100%). Another application was able to retrieve information from online publications and genomic databases to identify genes contributing to nonmedullary thyroid cancer, encompassing 95%−97% of all thyroid cancers.…”

“…In addition, NLP models were able to build large-volume data sets by extracting features from different free text sources, including diagnostic and pathology reports and clinician documentation. 12 , 21 Furthermore, some models standardized unstructured data from multiple institutions to generate multicenter data sets. 12 , 19 , 21 , 29 , 34 …”

“…Recent studies have demonstrated the effectiveness of various NLP systems in extracting relevant information from EHRs within the field of thyroidology. 12 These NLP-driven approaches have shown promise in enhancing and validating diagnostic and prognostic tools for various thyroid pathologies, including functional thyroid disorders, thyroid nodules, and thyroid cancer. 4 , 13 Despite these promising developments, the existing literature lacks a comprehensive overview.…”

OMOP CDM Can Facilitate Data-Driven Studies for Cancer Prediction: A Systematic Review

“…Clinical integration of the OMOP CDM can pave the way back to patients through facilitating access to relevant data, enabling multicentric, multidatabase studies to enhance statistical power and transfer results across populations [ 30 , 31 , 32 ]. With the recent advancements in the field of medical informatics, many predictive algorithms are known and used in the field of oncology, which benefits largely from the use of such models in uncovering unknown information about the cause and course of certain types of cancer.…”