Journal of healthcare informatics research最新文献

{"title":"BioBERTurk: Exploring Turkish Biomedical Language Model Development Strategies in Low-Resource Setting.","authors":"Hazal Türkmen, Oğuz Dikenelli, Cenk Eraslan, Mehmet Cem Çallı, Süha Süreyya Özbek","doi":"10.1007/s41666-023-00140-7","DOIUrl":"10.1007/s41666-023-00140-7","url":null,"abstract":"<p><p>Pretrained language models augmented with in-domain corpora show impressive results in biomedicine and clinical Natural Language Processing (NLP) tasks in English. However, there has been minimal work in low-resource languages. Although some pioneering works have shown promising results, many scenarios still need to be explored to engineer effective pretrained language models in biomedicine for low-resource settings. This study introduces the BioBERTurk family and four pretrained models in Turkish for biomedicine. To evaluate the models, we also introduced a labeled dataset to classify radiology reports of head CT examinations. Two parts of the reports, impressions and findings, are evaluated separately to observe the performance of models on longer and less informative text. We compared the models with the Turkish BERT (BERTurk) pretrained with general domain text, multilingual BERT (mBERT), and LSTM+attention-based baseline models. The first model initialized from BERTurk and then further pretrained with biomedical corpus performs statistically better than BERTurk, multilingual BERT, and baseline for both datasets. The second model continues to pretrain the BERTurk model by using only radiology Ph.D. theses to test the effect of task-related text. This model slightly outperformed all models on the impression dataset and showed that using only radiology-related data for continual pre-training could be effective. The third model continues to pretrain by adding radiology theses to the biomedical corpus but does not show a statistically meaningful difference for both datasets. The final model combines radiology and biomedicine corpora with the corpus of BERTurk and pretrains a BERT model from scratch. This model is the worst-performing model of the BioBERT family, even worse than BERTurk and multilingual BERT.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"7 4","pages":"433-446"},"PeriodicalIF":5.4,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10620363/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71490985","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Natural Language Processing to Classify Caregiver Strategies Supporting Participation Among Children and Youth with Craniofacial Microsomia and Other Childhood-Onset Disabilities.","authors":"Vera C Kaelin, Andrew D Boyd, Martha M Werler, Natalie Parde, Mary A Khetani","doi":"10.1007/s41666-023-00149-y","DOIUrl":"10.1007/s41666-023-00149-y","url":null,"abstract":"<p><p>Customizing participation-focused pediatric rehabilitation interventions is an important but also complex and potentially resource intensive process, which may benefit from automated and simplified steps. This research aimed at applying natural language processing to develop and identify a best performing predictive model that classifies caregiver strategies into participation-related constructs, while filtering out non-strategies. We created a dataset including 1,576 caregiver strategies obtained from 236 families of children and youth (11-17 years) with craniofacial microsomia or other childhood-onset disabilities. These strategies were annotated to four participation-related constructs and a non-strategy class. We experimented with manually created features (i.e., speech and dependency tags, predefined likely sets of words, dense lexicon features (i.e., Unified Medical Language System (UMLS) concepts)) and three classical methods (i.e., logistic regression, naïve Bayes, support vector machines (SVM)). We tested a series of binary and multinomial classification tasks applying 10-fold cross-validation on the training set (80%) to test the best performing model on the held-out test set (20%). SVM using term frequency-inverse document frequency (TF-IDF) was the best performing model for all four classification tasks, with accuracy ranging from 78.10 to 94.92% and a macro-averaged F1-score ranging from 0.58 to 0.83. Manually created features only increased model performance when filtering out non-strategies. Results suggest pipelined classification tasks (i.e., filtering out non-strategies; classification into intrinsic and extrinsic strategies; classification into participation-related constructs) for implementation into participation-focused pediatric rehabilitation interventions like Participation and Environment Measure Plus (PEM+) among caregivers who complete the Participation and Environment Measure for Children and Youth (PEM-CY).</p><p><strong>Supplementary information: </strong>The online version contains supplementary material available at 10.1007/s41666-023-00149-y.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"7 4","pages":"480-500"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10620347/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71490987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sequence Labeling for Disambiguating Medical Abbreviations.","authors":"Mucahit Cevik, Sanaz Mohammad Jafari, Mitchell Myers, Savas Yildirim","doi":"10.1007/s41666-023-00146-1","DOIUrl":"10.1007/s41666-023-00146-1","url":null,"abstract":"<p><p>Abbreviations are unavoidable yet critical parts of the medical text. Using abbreviations, especially in clinical patient notes, can save time and space, protect sensitive information, and help avoid repetitions. However, most abbreviations might have multiple senses, and the lack of a standardized mapping system makes disambiguating abbreviations a difficult and time-consuming task. The main objective of this study is to examine the feasibility of sequence labeling methods for medical abbreviation disambiguation. Specifically, we explore the capability of sequence labeling methods to deal with multiple unique abbreviations in a single text. We use two public datasets to compare and contrast the performance of several transformer models pre-trained on different scientific and medical corpora. Our proposed sequence labeling approach outperforms the more commonly used text classification models for the abbreviation disambiguation task. In particular, the SciBERT model shows a strong performance for both sequence labeling and text classification tasks over the two considered datasets. Furthermore, we find that abbreviation disambiguation performance for the text classification models becomes comparable to that of sequence labeling only when postprocessing is applied to their predictions, which involves filtering possible labels for an abbreviation based on the training data.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"7 4","pages":"501-526"},"PeriodicalIF":5.4,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10620358/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71490988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"State-of-the-Art of Breast Cancer Diagnosis in Medical Images via Convolutional Neural Networks (CNNs).","authors":"Pratibha Harrison, Rakib Hasan, Kihan Park","doi":"10.1007/s41666-023-00144-3","DOIUrl":"10.1007/s41666-023-00144-3","url":null,"abstract":"<p><p>Early detection of breast cancer is crucial for a better prognosis. Various studies have been conducted where tumor lesions are detected and localized on images. This is a narrative review where the studies reviewed are related to five different image modalities: histopathological, mammogram, magnetic resonance imaging (MRI), ultrasound, and computed tomography (CT) images, making it different from other review studies where fewer image modalities are reviewed. The goal is to have the necessary information, such as pre-processing techniques and CNN-based diagnosis techniques for the five modalities, readily available in one place for future studies. Each modality has pros and cons, such as mammograms might give a high false positive rate for radiographically dense breasts, while ultrasounds with low soft tissue contrast result in early-stage false detection, and MRI provides a three-dimensional volumetric image, but it is expensive and cannot be used as a routine test. Various studies were manually reviewed using particular inclusion and exclusion criteria; as a result, 91 recent studies that classify and detect tumor lesions on breast cancer images from 2017 to 2022 related to the five image modalities were included. For histopathological images, the maximum accuracy achieved was around 99 <math><mo>%</mo></math>, and the maximum sensitivity achieved was 97.29 <math><mo>%</mo></math> by using DenseNet, ResNet34, and ResNet50 architecture. For mammogram images, the maximum accuracy achieved was 96.52 <math><mo>%</mo></math> using a customized CNN architecture. For MRI, the maximum accuracy achieved was 98.33 <math><mo>%</mo></math> using customized CNN architecture. For ultrasound, the maximum accuracy achieved was around 99 <math><mo>%</mo></math> by using DarkNet-53, ResNet-50, G-CNN, and VGG. For CT, the maximum sensitivity achieved was 96 <math><mo>%</mo></math> by using Xception architecture. Histopathological and ultrasound images achieved higher accuracy of around 99 <math><mo>%</mo></math> by using ResNet34, ResNet50, DarkNet-53, G-CNN, and VGG compared to other modalities for either of the following reasons: use of pre-trained architectures with pre-processing techniques, use of modified architectures with pre-processing techniques, use of two-stage CNN, and higher number of studies available for Artificial Intelligence (AI)/machine learning (ML) researchers to reference. One of the gaps we found is that only a single image modality is used for CNN-based diagnosis; in the future, a multiple image modality approach can be used to design a CNN architecture with higher accuracy.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"7 4","pages":"387-432"},"PeriodicalIF":5.4,"publicationDate":"2023-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10620373/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71490989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Exploratory Study of the Readiness of Public Healthcare Facilities in Developing Countries to Adopt Health Information Technology (HIT)/e-Health: the Case of Ghana.","authors":"Salifu Yusif, Abdul Hafeez-Baig, Jeffrey Soar","doi":"10.1007/s41666-020-00070-8","DOIUrl":"https://doi.org/10.1007/s41666-020-00070-8","url":null,"abstract":"<p><p>There are myriad of factors used in assessing health information technology (HIT)/e-Health of healthcare institutions in developing countries and beyond. In this paper, we intended to identify and gain a deeper understanding of factors used in assessing HIT/e-Health readiness in developing countries through the identification of contextual attributes using Ghana as an exemplary developing country. Through in-depth interviews using <i>aide memoire</i> as interview guide, we explored <i>Core readiness</i>, <i>Engagement readiness</i>, <i>Technological readiness</i>, <i>HIT funding readiness</i>, <i>Regulatory and policy readiness</i>, <i>Workforce readiness and Change Management readiness.</i> We adapted the systematic thematic analysis of qualitative data guide suggested by Braun and Clarke (2013) and O'Connor and Gibson (Pimatisiwin 1: 63-90, 2003) in order to generate codes and build over-arching themes. While <i>Organizational cultural readiness</i> was found to be a more applicable theme/factor in place of <i>Engagement readiness</i> and <i>Change management readiness, Resource readiness</i> wasalso deemed a more appropriate theme for <i>HIT funding readiness</i> and <i>Workforce readiness</i> respectively. A total of 23 factors likely to promote HIT adoption in Ghana and 29 factors capable of impeding HIT adoption in Ghana and potentially in other developing countries were identified. For effective assessment of HIT readiness factors, there is a critical need for a deeper understanding of their applicability in differing settings. The outcome of this study offers a valuable insight into improving circumstances under which HIT/e-Health is adopted. When effectually carried out, assessment of this nature could be help side-step losses on large money, effort, time, delay and importantly, dissatisfaction among stakeholders while enabling change processes healthcare institutions and communities involved. This study also contributes to the limited literature on HIT/e-Health implementation scenarios while offering basis for theory-building.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"4 2","pages":"189-214"},"PeriodicalIF":5.4,"publicationDate":"2020-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8982759/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142057791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Individualized Modeling to Distinguish Between High and Low Arousal States Using Physiological Data.","authors":"Ame Osotsi, Zita Oravecz, Qunhua Li, Joshua Smyth, Timothy R Brick","doi":"10.1007/s41666-019-00064-1","DOIUrl":"10.1007/s41666-019-00064-1","url":null,"abstract":"<p><p>With wearable, relatively unobtrusive health monitors and smartphone sensors, it is increasingly easy to collect continuously streaming physiological data in a passive mode without placing much burden on participants. At the same time, smartphones provide the ability to survey participants to provide \"ground-truth\" reporting on psychological states, although this comes at an increased cost in participant burden. In this paper, we examined how analytical approaches from the field of machine learning could allow us to distill the collected physiological data into actionable decision rules about each individual's psychological state, with the eventual goal of identifying important psychological states (e.g., risk moments) without the need for ongoing burdensome active assessment (e.g., self-report). As a first step towards this goal, we compared two methods: (1) a k-nearest neighbor classifier that uses dynamic time warping distance, and (2) a random forests classifier to predict low and high states of affective arousal states based on features extracted using the <i>tsfresh</i> python package. Then, we compared random-forest-based predictive models tailored for the individual with individual-general models. Results showed that the individual-specific model outperformed the general one. Our results support the feasibility of using passively collected wearable data to predict psychological states, suggesting that by relying on both types of data, the active collection can be reduced or eliminated.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"4 1","pages":"91-109"},"PeriodicalIF":4.6,"publicationDate":"2020-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8982753/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144755592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting Glycaemia in Type 1 Diabetes Patients: Experiments in Feature Engineering and Data Imputation.","authors":"Jouhyun Jeon, Peter J Leimbigler, Gaurav Baruah, Michael H Li, Yan Fossat, Alfred J Whitehead","doi":"10.1007/s41666-019-00063-2","DOIUrl":"https://doi.org/10.1007/s41666-019-00063-2","url":null,"abstract":"<p><p>Patients with type 1 diabetes manually regulate blood glucose concentration by adjusting insulin dosage in response to factors such as carbohydrate intake and exercise intensity. Automated near-term prediction of blood glucose concentration is essential to prevent hyper- and hypoglycaemic events in type 1 diabetes patients and to improve control of blood glucose levels by physicians and patients. The imperfect nature of patient monitoring introduces missing values into all variables that play important roles to predict blood glucose level, necessitating data imputation. In this paper, we investigated the importance of variables and explored various feature engineering methods to predict blood glucose level. Next, we extended our work by developing a new empirical imputation method and investigating the predictive accuracy achieved under different methods to impute missing data. Also, we examined the influence of past signal values on the prediction of blood glucose levels. We reported the relative performance of predictive models in different testing scenarios and different imputation methods. Finally, we found an optimal combination of data imputation methods and built an ensemble model for the reliable prediction of blood glucose levels on a 30-minute horizon.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"4 1","pages":"71-90"},"PeriodicalIF":4.6,"publicationDate":"2019-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8982822/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149121268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detection of Surgical Site Infection Utilizing Automated Feature Generation in Clinical Notes.","authors":"Feichen Shen, David W Larson, James M Naessens, Elizabeth B Habermann, Hongfang Liu, Sunghwan Sohn","doi":"10.1007/s41666-018-0042-9","DOIUrl":"10.1007/s41666-018-0042-9","url":null,"abstract":"<p><p>Postsurgical complications (PSCs) are known as a deviation from the normal postsurgical course and categorized by severity and treatment requirements. Surgical site infection (SSI) is one of major PSCs and the most common healthcare-associated infection, resulting in increased length of hospital stay and cost. In this work, we proposed an automated way to generate keyword features using sublanguage analysis with heuristics to detect SSI from cohort in clinical notes and evaluated these keywords with medical experts. To further valid our approach, we also applied different machine learning algorithms on cohort using automatically generated keywords. The results showed that our approach was able to identify SSI keywords from clinical narratives and can be used as a foundation to develop an information extraction system or support search-based natural language processing (NLP) approaches by augmenting search queries.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"3 3","pages":"267-282"},"PeriodicalIF":4.6,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6855398/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Emotional Awareness and Decision-Making in the Context of Computer-Mediated Psychotherapy.","authors":"Ebrahim Oshni Alvandi, George Van Doorn, Mark Symmons","doi":"10.1007/s41666-019-00050-7","DOIUrl":"https://doi.org/10.1007/s41666-019-00050-7","url":null,"abstract":"<p><p>Emotional awareness has been previously investigated among clinicians. In this work, we bring to the fore of research the interest to uncover emotional awareness of clinicians during the tele-mental health session. The study reported here aimed at determining whether clinicians process their own emotions, as well as those of the client, in a computer-mediated context. Also, clinicians' decision-making process was assessed because such action appears to be related to the way they feel and recognise how those emotions may change their thinking and impact their interaction with clients. We estimated that such ability in clinicians' would be contrasted when the psychotherapy-session level is conducted via various technologies. Participant of the study were presented by stimuli in different modes of delivery (e.g. text, audio, and video). The experiment indicates that the ability to manage, perceive, and utilise emotions was as being satisfactory during all modes of delivery. In essence, the findings contribute to the field of remote therapy suggesting emotional awareness as a key cognitive factor in diagnosis.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"3 3","pages":"345-370"},"PeriodicalIF":5.4,"publicationDate":"2019-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8982805/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142057789","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Discovering Oculometric Patterns to Detect Cognitive Performance Changes in Healthy Youth Football Athletes.","authors":"Gaurav N Pradhan, Jamie M Bogle, Michael J Cevette, Jan Stepanek","doi":"10.1007/s41666-019-00045-4","DOIUrl":"https://doi.org/10.1007/s41666-019-00045-4","url":null,"abstract":"<p><p>In this paper, we focus on the application of oculometric patterns extracted from raw eye movements during a mental workload task to assess changes in cognitive performance in healthy youth athletes over the course of a typical sport season. Oculometric features pertaining to fixations and saccades were measured on 116 athletes in pre- and post-season testing. Participants were between 7 and 14 years of age at pre-season testing. Due to varied developmental rates, there were large interindividual performance differences during a mental workload task consisting of reading numbers. Based on different reading speeds, we classified three profiles (slow, moderate, and fast) and established their corresponding baselines for oculometric data. Within each profile, we describe changes in oculomotor function based on changes in cognitive performance during the season. To visualize these changes in multidimensional oculometric data, we also present a multidimensional visualization tool named DiViTo (diagnostic visualization tool). These experimental, computational informatics and visualization methodologies may serve to utilize oculometric information to detect changes in cognitive performance due to mild or severe cognitive impairment such as concussion/mild traumatic brain injury, as well as possibly other disorders such as attention deficit hyperactivity disorders, learning/reading disabilities, impairment of alertness, and neurocognitive function.</p>","PeriodicalId":101413,"journal":{"name":"Journal of healthcare informatics research","volume":"3 4","pages":"371-392"},"PeriodicalIF":5.4,"publicationDate":"2019-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8982780/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142057790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}