Automated medical chart review for breast cancer outcomes research: a novel natural language processing extraction system.

BACKGROUND: Manually extracted data points from health records are collated on an institutional, provincial, and national level to facilitate clinical research. However, the labour-intensive clinical chart review process puts an increasing burden on healthcare system budgets. Therefore, an automated information extraction system is needed to ensure the timeliness and scalability of research data. METHODS: We used a dataset of 100 synoptic operative and 100 pathology reports, evenly split into 50 reports in training and test sets for each report type. The training set guided our development of a Natural Language Processing (NLP) extraction pipeline system, which accepts scanned images of operative and pathology reports. The system uses a combination of rule-based and transfer learning methods to extract numeric encodings from text. We also developed visualization tools to compare the manual and automated extractions. The code for this paper was made available on GitHub. RESULTS: A test set of 50 operative and 50 pathology reports were used to evaluate the extraction accuracies of the NLP pipeline. Gold standard, defined as manual extraction by expert reviewers, yielded accuracies of 90.5% for operative reports and 96.0% for pathology reports, while the NLP system achieved overall 91.9% (operative) and 95.4% (pathology) accuracy. The pipeline successfully extracted outcomes data pertinent to breast cancer tumor characteristics (e.g. presence of invasive carcinoma, size, histologic type), prognostic factors (e.g. number of lymph nodes with micro-metastases and macro-metastases, pathologic stage), and treatment-related variables (e.g. margins, neo-adjuvant treatment, surgical indication) with high accuracy. Out of the 48 variables across operative and pathology codebooks, NLP yielded 43 variables with F-scores of at least 0.90; in comparison, a trained human annotator yielded 44 variables with F-scores of at least 0.90. CONCLUSIONS: The NLP system achieves near-human-level accuracy in both operative and pathology reports using a minimal curated dataset. This system uniquely provides a robust solution for transparent, adaptable, and scalable automation of data extraction from patient health records. It may serve to advance breast cancer clinical research by facilitating collection of vast amounts of valuable health data at a population level.

FGF21 does not require adipocyte AMP-activated protein kinase (AMPK) or the phosphorylation of acetyl-CoA carboxylase (ACC) to mediate improvements in whole-body glucose homeostasis.

OBJECTIVE: Fibroblast growth factor 21 (FGF21) shows great potential for the treatment of obesity and type 2 diabetes, as its long-acting analogue reduces body weight and improves lipid profiles of participants in clinical studies; however, the intracellular mechanisms mediating these effects are poorly understood. AMP-activated protein kinase (AMPK) is an important energy sensor of the cell and a molecular target for anti-diabetic medications. This work examined the role of AMPK in mediating the glucose and lipid-lowering effects of FGF21. METHODS: Inducible adipocyte AMPK ß1ß2 knockout mice (iß1ß2AKO) and littermate controls were fed a high fat diet (HFD) and treated with native FGF21 or saline for two weeks. Additionally, HFD-fed mice with knock-in mutations on the AMPK phosphorylation sites of acetyl-CoA carboxylase (ACC)1 and ACC2 (DKI mice) along with wild-type (WT) controls received long-acting FGF21 for two weeks. RESULTS: Consistent with previous studies, FGF21 treatment significantly reduced body weight, adiposity, and liver lipids in HFD fed mice. To add, FGF21 improved circulating lipids, glycemic control, and insulin sensitivity. These effects were independent of adipocyte AMPK and were not associated with changes in browning of white (WAT) and brown adipose tissue (BAT). Lastly, we assessed whether FGF21 exerted its effects through the AMPK/ACC axis, which is critical in the therapeutic benefits of the anti-diabetic medication metformin. ACC DKI mice had improved glucose and insulin tolerance and a reduction in body weight, body fat and hepatic steatosis similar to WT mice in response to FGF21 administration. CONCLUSIONS: These data illustrate that the metabolic improvements upon FGF21 administration are independent of adipocyte AMPK, and do not require the inhibitory action of AMPK on ACC. This is in contrast to the anti-diabetic medication metformin and suggests that the treatment of obesity and diabetes with the combination of FGF21 and AMPK activators merits consideration.