The augmented value of using clinical notes in semi-automated surveillance of deep surgical site infections after colorectal surgery.

BACKGROUND: In patients who underwent colorectal surgery, an existing semi-automated surveillance algorithm based on structured data achieves high sensitivity in detecting deep surgical site infections (SSI), however, generates a significant number of false positives. The inclusion of unstructured, clinical narratives to the algorithm may decrease the number of patients requiring manual chart review. The aim of this study was to investigate the performance of this semi-automated surveillance algorithm augmented with a natural language processing (NLP) component to improve positive predictive value (PPV) and thus workload reduction (WR). METHODS: Retrospective, observational cohort study in patients who underwent colorectal surgery from January 1, 2015, through September 30, 2020. NLP was used to detect keyword counts in clinical notes. Several NLP-algorithms were developed with different count input types and classifiers, and added as component to the original semi-automated algorithm. Traditional manual surveillance was compared with the NLP-augmented surveillance algorithms and sensitivity, specificity, PPV and WR were calculated. RESULTS: From the NLP-augmented models, the decision tree models with discretized counts or binary counts had the best performance (sensitivity 95.1% (95%CI 83.5-99.4%), WR 60.9%) and improved PPV and WR by only 2.6% and 3.6%, respectively, compared to the original algorithm. CONCLUSIONS: The addition of an NLP component to the existing algorithm had modest effect on WR (decrease of 1.4-12.5%), at the cost of sensitivity. For future implementation it will be a trade-off between optimal case-finding techniques versus practical considerations such as acceptability and availability of resources.

Explainable AI for Data-Driven Feedback and Intelligent Action Recommendations to Support Students Self-Regulation.

Formative feedback has long been recognised as an effective tool for student learning, and researchers have investigated the subject for decades. However, the actual implementation of formative feedback practices is associated with significant challenges because it is highly time-consuming for teachers to analyse students' behaviours and to formulate and deliver effective feedback and action recommendations to support students' regulation of learning. This paper proposes a novel approach that employs learning analytics techniques combined with explainable machine learning to provide automatic and intelligent feedback and action recommendations that support student's self-regulation in a data-driven manner, aiming to improve their performance in courses. Prior studies within the field of learning analytics have predicted students' performance and have used the prediction status as feedback without explaining the reasons behind the prediction. Our proposed method, which has been developed based on LMS data from a university course, extends this approach by explaining the root causes of the predictions and by automatically providing data-driven intelligent recommendations for action. Based on the proposed explainable machine learning-based approach, a dashboard that provides data-driven feedback and intelligent course action recommendations to students is developed, tested and evaluated. Based on such an evaluation, we identify and discuss the utility and limitations of the developed dashboard. According to the findings of the conducted evaluation, the dashboard improved students' learning outcomes, assisted them in self-regulation and had a positive effect on their motivation.

Using machine learning for predicting cervical cancer from Swedish electronic health records by mining hierarchical representations.

Electronic health records (EHRs) contain rich documentation regarding disease symptoms and progression, but EHR data is challenging to use for diagnosis prediction due to its high dimensionality, relative scarcity, and substantial level of noise. We investigated how to best represent EHR data for predicting cervical cancer, a serious disease where early detection is beneficial for the outcome of treatment. A case group of 1321 patients with cervical cancer were matched to ten times as many controls, and for both groups several types of events were extracted from their EHRs. These events included clinical codes, lab results, and contents of free text notes retrieved using a LSTM neural network. Clinical events are described with great variation in EHR texts, leading to a very large feature space. Therefore, an event hierarchy inferred from the textual events was created to represent the clinical texts. Overall, the events extracted from free text notes contributed the most to the final prediction, and the hierarchy of textual events further improved performance. Four classifiers were evaluated for predicting a future cancer diagnosis where Random Forest achieved the best results with an AUC of 0.70 from a year before diagnosis up to 0.97 one day before diagnosis. We conclude that our approach is sound and had excellent discrimination at diagnosis, but only modest discrimination capacity before this point. Since our study objective was earlier disease prediction than such, we propose further work should consider extending patient histories through e.g. the integration of primary health records preceding referral to hospital.

Recent advances in Swedish and Spanish medical entity recognition in clinical texts using deep neural approaches.

BACKGROUND: Text mining and natural language processing of clinical text, such as notes from electronic health records, requires specific consideration of the specialized characteristics of these texts. Deep learning methods could potentially mitigate domain specific challenges such as limited access to in-domain tools and data sets. METHODS: A bi-directional Long Short-Term Memory network is applied to clinical notes in Spanish and Swedish for the task of medical named entity recognition. Several types of embeddings, both generated from in-domain and out-of-domain text corpora, and a number of generation and combination strategies for embeddings have been evaluated in order to investigate different input representations and the influence of domain on the final results. RESULTS: For Spanish, a micro averaged F1-score of 75.25 was obtained and for Swedish, the corresponding score was 76.04. The best results for both languages were achieved using embeddings generated from in-domain corpora extracted from electronic health records, but embeddings generated from related domains were also found to be beneficial. CONCLUSIONS: A recurrent neural network with in-domain embeddings improved the medical named entity recognition compared to shallow learning methods, showing this combination to be suitable for entity recognition in clinical text for both languages.

Semi-supervised medical entity recognition: A study on Spanish and Swedish clinical corpora.

OBJECTIVE: The goal of this study is to investigate entity recognition within Electronic Health Records (EHRs) focusing on Spanish and Swedish. Of particular importance is a robust representation of the entities. In our case, we utilized unsupervised methods to generate such representations. METHODS: The significance of this work stands on its experimental layout. The experiments were carried out under the same conditions for both languages. Several classification approaches were explored: maximum probability, CRF, Perceptron and SVM. The classifiers were enhanced by means of ensembles of semantic spaces and ensembles of Brown trees. In order to mitigate sparsity of data, without a significant increase in the dimension of the decision space, we propose the use of clustered approaches of the hierarchical Brown clustering represented by trees and vector quantization for each semantic space. RESULTS: The results showed that the semi-supervised approaches significantly improved standard supervised techniques for both languages. Moreover, clustering the semantic spaces contributed to the quality of the entity recognition while keeping the dimension of the feature-space two orders of magnitude lower than when directly using the semantic spaces. CONCLUSIONS: The contributions of this study are: (a) a set of thorough experiments that enable comparisons regarding the influence of different types of features on different classifiers, exploring two languages other than English; and (b) the use of ensembles of clusters of Brown trees and semantic spaces on EHRs to tackle the problem of scarcity of available annotated data.

Finding Cervical Cancer Symptoms in Swedish Clinical Text using a Machine Learning Approach and NegEx.

Detection of early symptoms in cervical cancer is crucial for early treatment and survival. To find symptoms of cervical cancer in clinical text, Named Entity Recognition is needed. In this paper the Clinical Entity Finder, a machine-learning tool trained on annotated clinical text from a Swedish internal medicine emergency unit, is evaluated on cervical cancer records. The Clinical Entity Finder identifies entities of the types body part, finding and disorder and is extended with negation detection using the rule-based tool NegEx, to distinguish between negated and non-negated entities. To measure the performance of the tools on this new domain, two physicians annotated a set of clinical notes from the health records of cervical cancer patients. The inter-annotator agreement for finding, disorder and body part obtained an average F-score of 0.677 and the Clinical Entity Finder extended with NegEx had an average F-score of 0.667.