Deep Learning-based detection of psychiatric attributes from German mental health records

Madan, Sumit; Zimmer, Fabian Julius; Balabin, Helena; Schaaf, Sebastian; Fröhlich, Holger; Fluck, Juliane; Neuner, Irene; Mathiak, Klaus; Hofmann‐Apitius, Martin; Sarkheil, Pegah

doi:10.1016/j.ijmedinf.2022.104724

Cited by 7 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The rule-based methods mostly extract symptoms through keyword retrieval, which ignores the context semantics of each symptom. Recently, Bert (35) was applied to various clinical tasks (50)(51)(52). Through a multi-head attention mechanism, Bert-related models can obtain context features of each symptom.…”

Section: Discussionmentioning

confidence: 99%

Association between depressive symptoms and diagnosis of diabetes and its complications: A network analysis in electronic health records

Wan

Feng²,

Ma³

et al. 2022

Front. Psychiatry

View full text Add to dashboard Cite

ObjectivesDiabetes and its complications are commonly associated with depressive symptoms, and few studies have investigated the diagnosis effect of depressive symptoms in patients with diabetes. The present study used a network-based approach to explore the association between depressive symptoms, which are annotated from electronic health record (EHR) notes by a deep learning model, and the diagnosis of type 2 diabetes mellitus (T2DM) and its complications.MethodsIn this study, we used anonymous admission notes of 52,139 inpatients diagnosed with T2DM at the first affiliated hospital of Nanjing Medical University from 2008 to 2016 as input for a symptom annotation model named T5-depression based on transformer architecture which helps to annotate depressive symptoms from present illness. We measured the performance of the model by using the F1 score and the area under the receiver operating characteristic curve (AUROC). We constructed networks of depressive symptoms to examine the connectivity of these networks in patients diagnosed with T2DM, including those with certain complications.ResultsThe T5-depression model achieved the best performance with an F1-score of 91.71 and an AUROC of 96.25 compared with the benchmark models. The connectivity of depressive symptoms in patients diagnosed with T2DM (p = 0.025) and hypertension (p = 0.013) showed a statistically significant increase 2 years after the diagnosis, which is consistent with the number of patients diagnosed with depression.ConclusionThe T5-depression model proposed in this study can effectively annotate depressive symptoms in EHR notes. The connectivity of annotated depressive symptoms is associated with the diagnosis of T2DM and hypertension. The changes in the network of depressive symptoms generated by the T5-depression model could be used as an indicator for screening depression.

show abstract

Section: Discussionmentioning

confidence: 99%

Association between depressive symptoms and diagnosis of diabetes and its complications: A network analysis in electronic health records

Wan

Feng²,

Ma³

et al. 2022

Front. Psychiatry

View full text Add to dashboard Cite

show abstract

“…That same year, three studies were published that utilized the bidirectional encoder representations from transformers (BERT) architecture [28,43,46]. By 2021 [47][48][49][50][51][52][53][54], the BERT architecture and its variants had emerged as the primary NER model applied to EHRs, a trend that continues to this day [4,24,25,[55][56][57][58][59][60][61][62][63][64][65][66][67][68][69][70][71]. However, this self-attention mechanism was initially introduced in 2017 [72].…”

Section: Classification Modelsmentioning

confidence: 99%

Named Entity Recognition in Electronic Health Records: A Methodological Review

Durango,

Torres-Silva,

Orozco-Duque

2023

Healthc Inform Res

View full text Add to dashboard Cite

Objectives: A substantial portion of the data contained in Electronic Health Records (EHR) is unstructured, often appearing as free text. This format restricts its potential utility in clinical decision-making. Named entity recognition (NER) methods address the challenge of extracting pertinent information from unstructured text. The aim of this study was to outline the current NER methods and trace their evolution from 2011 to 2022.Methods: We conducted a methodological literature review of NER methods, with a focus on distinguishing the classification models, the types of tagging systems, and the languages employed in various corpora.Results: Several methods have been documented for automatically extracting relevant information from EHRs using natural language processing techniques such as NER and relation extraction (RE). These methods can automatically extract concepts, events, attributes, and other data, as well as the relationships between them. Most NER studies conducted thus far have utilized corpora in English or Chinese. Additionally, the bidirectional encoder representation from transformers using the BIO tagging system architecture is the most frequently reported classification scheme. We discovered a limited number of papers on the implementation of NER or RE tasks in EHRs within a specific clinical domain.Conclusions: EHRs play a pivotal role in gathering clinical information and could serve as the primary source for automated clinical decision support systems. However, the creation of new corpora from EHRs in specific clinical domains is essential to facilitate the swift development of NER and RE models applied to EHRs for use in clinical practice.

show abstract

“…According to Madan et al, [14], there are vast real-world databases of health care records available worldwide that can be used to improve personalized medicine and predictive analysis. Of these, the major source of information collection in mental health has been text-based records.…”

Section: Related Workmentioning

confidence: 99%

Predictive Analysis of Mental Health Conditions Using AdaBoost Algorithm

Ogunseye

Adenusi²,

Nwanakwaugwu

et al. 2022

paradigmplus

View full text Add to dashboard Cite

The presented research responds to increased mental illness conditions worldwide and the need for efficient mental health care (MHC) through machine learning (ML) implementations. The datasets employed in this investigation belong to a Kaggle repository named "Mental Health Tech Survey." The surveys for the years 2014 and 2016 were downloaded and aggregated. The prediction results for bagging, stacking, LR, KNN, tree class, NN, RF, and Adaboost yielded 75.93%, 75.93%, 79.89%, 90.42%, 80.69%, 89.95%, 81.22%, and 81.75% respectively. The AdaBoost ML model performed data cleaning and prediction on the datasets, reaching an accuracy of 81.75%, which is good enough for decision-making. The results were further used with other ML models such as Random Forest (RF), K-Nearest Neighbor (KNN), bagging, and a few others, with reported accuracy ranging from 81.22 to 75.93 which is good enough for decision making. Out of all the models used for predicting mental health treatment outcomes, AdaBoost has the highest accuracy.

show abstract

Deep Learning-based detection of psychiatric attributes from German mental health records

Cited by 7 publications

References 12 publications

Association between depressive symptoms and diagnosis of diabetes and its complications: A network analysis in electronic health records

Association between depressive symptoms and diagnosis of diabetes and its complications: A network analysis in electronic health records

Named Entity Recognition in Electronic Health Records: A Methodological Review

Predictive Analysis of Mental Health Conditions Using AdaBoost Algorithm

Contact Info

Product

Resources

About