Background With a motivation of quality assurance, machine learning techniques were trained to classify Norwegian radiology reports of paediatric CT examinations according to their description of abnormal findings. Methods 13.506 reports from CT-scans of children, 1000 reports from CT scan of adults and 1000 reports from X-ray examination of adults were classified as positive or negative by a radiologist, according to the presence of abnormal findings. Inter-rater reliability was evaluated by comparison with a clinician’s classifications of 500 reports. Test–retest reliability of the radiologist was performed on the same 500 reports. A convolutional neural network model (CNN), a bidirectional recurrent neural network model (bi-LSTM) and a support vector machine model (SVM) were trained on a random selection of the children’s data set. Models were evaluated on the remaining CT-children reports and the adult data sets. Results Test–retest reliability: Cohen’s Kappa = 0.86 and F1 = 0.919. Inter-rater reliability: Kappa = 0.80 and F1 = 0.885. Model performances on the Children-CT data were as follows. CNN: (AUC = 0.981, F1 = 0.930), bi-LSTM: (AUC = 0.978, F1 = 0.927), SVM: (AUC = 0.975, F1 = 0.912). On the adult data sets, the models had AUC around 0.95 and F1 around 0.91. Conclusions The models performed close to perfectly on its defined domain, and also performed convincingly on reports pertaining to a different patient group and a different modality. The models were deemed suitable for classifying radiology reports for future quality assurance purposes, where the fraction of the examinations with abnormal findings for different sub-groups of patients is a parameter of interest.
Infections caused by central venous catheter (CVC) use is a serious and under-reported problem. In our research we explore methods of automatically detecting CVC use from clinical documentation for quality improvement and surveillance purposes. This paper describes our initial research on this topic, where we build CVC event classifiers based on an episodes of care corpus and an annotated gold standard. After describing the available data and gold standard we then experiment with different classification algorithms and feature selection approaches. We find that even with limited data it is possible to build reasonably accurate sentence classifiers, at least for the events that are most important to us. We also find that making use of document meta information may help improve classification quality by providing additional context to a sentence. Finally, we outline some strategies on using these preliminary clinical document-centric results as a tool for future analysis and elicitation of CVC usage intervals over full patient histories.
Background Treatment options in patients with advanced urothelial carcinoma (aUC) have improved significantly in the last decade, highlighting the need for adequate biomarkers. High tumor CGB5 mRNA, encoding Human Chorionic Gonadotrophin β (hCGβ) and elevated serum free hCGβ are related to poorly differentiated UC and correlated to poor prognosis. Hypothetically, hCG could be used as serum tumor marker in aUC patients. Patients and methods From December 2018 to November 2020, serum hCG levels were measured in 62 aUC patients referred to the oncology department, Akershus University Hospital. Median age 68 years, 42 male and 21 female. Total serum hCG (intact hCG + free hCGβ) was measured by electrochemiluminescence immunoassay “ECLIA”. In order to compensate for hypogonadism-related increases, hCG ≥ 50% above upper normal limit was considered as being elevated (hCG+), otherwise it was considered negative (hCG-). Further, hCG levels were defined as increasing (↑) or declining (↓) at treatment evaluation and follow up. Radiologically, patients had newly diagnosed disease (ND), progressive disease (PD) or non-PD (stable disease, partial or complete response) according to RECIST 1.1 at evaluation and follow up. hCG values were correlated with radiological findings at diagnosis, during and after treatment. Statistics by 2 x 2 contingency table and Chi Square (χ2) test. Results At least one hCG+ value was measured in 38 of 62 patients (61%), range 0.8 - 62,700 IU/L, and hCG was elevated in 22 of 43 patients (51%) with available hCG value at diagnosis, mean 15 IU/L (range 0.8 - 58). In total 213 hCG measurements could be correlated with concomitant CT or MR scans. At radiologic evaluation of ND & PD, hCG+/↑ and hCG -/↓ was found in 62 and 50 measurements, respectively. At non-PD, hCG -/↓ and hCG+/↑ was found in 94 and 7 measurements, respectively. For ND & PD versus non-PD, hCG showed a sensitivity and specificity of 55% and 93% respectively (95% CI 46-65% and 86-97%), positive and negative predictive value of 90% and 65% (95% CI 81-95% and 60-70%) and accuracy of 73% (95% CI 66-79%), χ2: p<0.00001. Exclusion of the 24 patients without elevated hCG showed improved sensitivity 76% at the cost of lower specificity 82% (95% CI 65-85% and 66-92% respectively), positive and negative predictive value 90% and 63% (95% CI 82-95% and 52-72%) and accuracy of 78% (95% CI 70-86%), χ2: p<0.00001. Conclusions Ectopic hCG production was observed in 61% of aUC patients with a strong and meaningful correlation between hCG changes and radiologic evaluation of treatment response. Intriguingly, hCG increase sometimes preceded radiologic PD, thus categorized as “false positive” until subsequent radiologic PD. hCG appears to be a promising tumor marker for biochemical treatment response evaluation in aUC patients. Citation Format: Katarina Puco, Johan Bjerner, Stig Müller, Gunder M. Lilleaasen, Haldor Husby, Mattias Røed-Undlien, Daniel Heinrich, Fredrik A. Dahl, Oluf D. Røe, Jan Oldenburg. Utility of hCG as a biomarker of treatment response in advanced urothelial cancer: A population based series at Akershus University Hospital, Norway [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 481.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.