Medical Health Record Research Articles

Navigating a "Good Death" During COVID-19: Understanding Real-Time End-of-Life Care Structures, Processes, and Outcomes Through Clinical Notes.

The coronavirus disease 2019 (COVID-19) pandemic severely disrupted hospice care, yet there is little research regarding how widespread disruptions affected clinician and family decision-making. We aimed to understand how the pandemic affected structures, processes, and outcomes of end-of-life care. Retrospective narrative chart review of electronic health records of 61 patients referred and admitted to hospice from 3 New York City geriatrics practices who died between March 1, 2020, and March 31, 2021. We linked longitudinal, unstructured medical, and hospice electronic health record notes to create a real-time, multiperspective trajectory of patients' interactions with providers using directed content analysis. Most patients had dementia and were enrolled in hospice for 11 days. Care processes were shaped by structural factors (staffing, supplies, and governmental/institutional policies), and outcomes were prioritized by care teams and families (protecting safety, maintaining high-touch care, honoring patient values, and supporting patients emotionally and spiritually). Processes used to achieve these outcomes were decision-making, care delivery, supporting a "good death," and emotional and spiritual support. Care processes were negotiated throughout the end of life, with clinicians and families making in-the-moment decisions. Some adaptations were effective but also placed extraordinary pressure on paid and family caregivers. Healthcare teams' and families' goals to meet patients' end-of-life priorities can be supported by ongoing assessment of patient goals and process changes needed to support them, stronger structural supports for paid and family caregivers, incentivizing relationships across primary care and hospice teams, and extending social work and spiritual care.

A Novel Fragmented Approach for Securing Medical Health Records in Multimodal Medical Images

A Novel Fragmented Approach for Securing Medical Health Records in Multimodal Medical Images

The Effects of Multidisciplinary Intensive Rehabilitation on Cognitive and Executive Functions in Parkinson's Disease: A Clinical Database Analysis.

Background/Objectives: This study is based on data collected from a medical health record review to assess whether multidisciplinary intensive rehabilitation treatment in Parkinson's disease (PD) patients can improve global cognitive functioning and executive functions. Methods: The data related to PD patients were extrapolated from a clinical database called "NeuroRehab". A total of 104 PD patients (51 males; 53 females) performed 6 weeks of multidisciplinary intensive rehabilitation treatment in clinical practice from January 2019 to May 2023. This training program was characterized by three daily sessions of 60 min of activities (muscle relaxation and stretching exercises, moderate physical aerobic exercise, and occupational therapy). The patients were classified and stratified according to disease severity (according to the Hoehn and Yahr scale), postural instability and gait difficulty (PIGD) or tremor-dominant (TD) subtypes, disease duration (DD), and the presence of dyskinesias. The effect of multidisciplinary intensive rehabilitation treatment on cognitive and executive functions was evaluated through the administration of cognitive tests, such as the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), and Frontal Assessment Battery (FAB). All the parameters were evaluated at the baseline (T0) and at the end of the rehabilitation program (T1). Results: The multidisciplinary intensive rehabilitation treatment significantly improved cognitive performance. The MMSE, MoCA, and FAB test scores after the rehabilitation program (T1) were significantly higher compared to the scores obtained at the baseline (T0). Moreover, further analyses on subgroups of the patients who scored below the cut-off in the MMSE showed that at least 50% of patients overcame the cut-off score. Interestingly, the same analyses performed for the MoCA and FAB revealed a higher rate of improvement in cognitive functions, with normal scores in both tests after 6 weeks of multidisciplinary intensive rehabilitation treatment. Conclusions: This study revealed the potential effects of a 6-week multidisciplinary rehabilitation program in improving cognitive status in a PD inpatient cohort.

NEUROOPHTHALMOLOGIC MANIFESTATIONS AMONG 195 PATIENTS WITH INTRACRANIAL TUMOR AT NATIONAL REFERRAL HOSPITAL

Introduction & Objective: Intracranial tumor might show neuroophthalmological manifestations. Early detection and prompt treatment could alter the visual outcome. Ophthalmologists need to be aware of neuroophthalmological features commonly found among patients with intracranial tumor. This study revealed those signs and symptoms. Methods: Study was done in tertiary referral hospital. Subjects were patients with intracranial tumor attending neuroophthalmology clinic between the specified period. It was a descriptive study using retrospective data of the patient’s initial visit to the clinic. Data was retrieved from the medical record archive and electronic health record. Results : There were 195 patients, 67.7% of them were women, with average age 43.4 years old. Half the patients arrived within 6 months after eye related symptoms occurred and the referring physicians was equal between ophthalmologists and other specialties. The most frequent symptom and sign were blurred vision and optic nerve head abnormality (atrophy, edema). Hemianopia outnumbered other patterns of visual field disturbance, yet bilateral pattern was difficult to be evaluated as most patients had either one or both eyes blind. Conclusion : Intracranial tumor was one of the diseases that could show ophthalmic signs and symptoms. Neuroophthalmological manifestations in those cases was related to the extraocular process and the relevant pathway affected by the tumor.

76 Lessons learned during implementation of OMOP common data model across multiple health systems

OBJECTIVES/GOALS: Adoption of the Observational Medical Outcomes Partnership (OMOP) common data model promises to transform large-scale observational health research. However, there are diverse challenges for operationalizing OMOP in terms of interoperability and technical skills among coordinating centers throughout the US. METHODS/STUDY POPULATION: A team from the Critical Path Institute (C-Path) collaborated with the informatics team members at Johns Hopkins to provide technical support to participating sites as part of the Extract, Transform, and Load (ETL) process linking existing concepts to OMOP concepts. Health systems met regularly via teleconference to review challenges and progress in ETL process. Sites were responsible for performing the local ETL process with assistance and securely provisioning de-identified data as part of the CURE ID program. RESULTS/ANTICIPATED RESULTS: More than twenty health systems participated in the CURE ID effort.Laboratory measures, basic demographics, disease diagnoses and problem list were more easily mapped to OMOP concepts by CURE ID partner institutions. Outcomes, social determinants of health, medical devices, and specific treatments were less easily characterized as part of the project. Concepts within the medical record presented very different technical challenges in terms of representation. There is a lack of standardization in OMOP implementation even among centers using the same electronic medical health record. Readiness to adopt OMOP varied across the institutions who participated. Health systems achieved variable level of coverage using OMOP medical concepts as part of the initiative. DISCUSSION/SIGNIFICANCE: Adoption of OMOP involves local stakeholder knowledge and implementation. Variable complexity of health concepts contributed to variable coverage. Documentation and support require extensive time and effort. Open-source software can be technically challenging. Interoperability of secure data systems presents unique problems.

Graph Neural Network Based Multi-Label Hierarchical Classification for Disease Predictions in General Practice.

General practitioners are supposed to be better diagnostics to detect patients with serious diseases earlier, and conduct early interventions and appropriate referrals of patients. However, in the current general practice, primary general practitioners lack sufficient clinical experiences, and the correct rate of general disease diagnosis is low. To assist general practitioners in diagnosis, this paper proposes a multi-label hierarchical classification method based on graph neural network, which integrates medical knowledge and electronic health record (EHR) data to build a disease prediction model. The experimental results based on data consist of 231,783 visits from EHR show that the proposed model outperforms all baseline models in the general disease prediction task with a top-3 recall of 0.865. The interpretable results of the model can effectively help clinicians understand the basis of the model's decision-making.

Tailoring community preventive care to personal health needs: a data-driven and participatory approach

A digital platform with tools for health promotion was developed as a proof of concept in co-creation with all relevant stakeholders involved: citizens, caregivers, and local policy makers, aiming to optimize community preventive care.
 The platform, referred to as BIBOPP, was developed as a proof of concept focusing on prevention and health promotion for citizens, local policy makers but also for primary care givers wanting to offer personalized follow up. To this end, we set up multiple co-creation sessions with different target groups of end users, including those with expected limited digital literacy skills, eg. older adults and people with no or low formal educational background. Based on this input, the functionalities of the tools were designed and developed and afterwards tested with a new group of end-users.
 The online prevention platform offered citizens from 5 Flemish municipalities a digital screening, allowing them to evaluate their health and lifestyle. This screening is the Domus Medica Health Guide (GGDM), an evidence-based tool predicting personal risk for chronic diseases. The outcome of the screening is linked to the electronic medical health record hosted by the general practitioner, notifying the latter in case of excessive risk and involving him in the process of disease prevention and health promotion.
 Based on the personalized lifestyle advice resulting from the survey responses, end users then received an overview of activities and primary care providers in their community, tailored to their personal risk profile and preferences.
 The data collected with consent from the end users, yielded insights on the self-reported lifestyle and health of these participants. On an aggregated level, this tool has the potential to provide data for population health management.
 First lessons learned indicate that the platform did require significant digital literacy skills from end-users, as indicated by the participants younger age and relatively healthy lifestyle (eg. less than 10% smokers, +50% daily physical activity, …). However, in a co-creation session group with older adults, they indicated that digital skills were not a threshold when peer support was available. Furthermore, they recognized that the tool helped them in the self-management of their health and that they trusted the data governance. People with low to no formal educational background on the other hand were more suspicious about the tool and data governance unless this was offered to them through their trusted care providers. Also, older adults indicated that support with the interpretation of the advices by a health professional could be of added value.
 Next steps for BIBOPP now are involving these health professionals, and more specifically primary care actors in onboarding and guiding people to and through the platform.
 To conclude, this data-driven proof of concept co-designed with end-users, has promise to deliver personalized services, strengthen community preventive care, and optimize the local health policy.
 

Electronic Health Record Tools to Facilitate Diagnosis of Cognitive Impairment in Primary Care

AbstractWith novel biomarkers and treatments for dementia on the horizon, the need to detect cognitive impairment at early stages is urgent. The role of primary care providers (PCPs) in the diagnosis of mild cognitive impairment and dementia is essential in meeting this need. Cognition in Primary Care (CPC) is a quality improvement program developed at the University of Washington (UW), Seattle, which includes a three‐part Continuing Medical Education (CME), clinical workflow, and electronic health record toolset to support PCPs in early diagnosis of mild cognitive impairment and dementia. This program was developed using input from UW providers in primary care practice and elements from the KAER (Kickstart, Assess, Evaluate, Refer) Toolkit developed by the Gerontological Society of America (GSA). Our pragmatic approach involved the development of tools integrated into Epic, the electronic health record system in place at UW, that can be used during a patient visit to guide PCPs through the process of evaluating cognitive function. The toolset includes 1) a list of signs that may trigger the need for a dedicated follow‐up visit to evaluate cognition; 2) a cognitive checklist for related conditions affecting cognition (e.g., potentially harmful medications, depressive symptoms, alcohol use, and obstructive sleep apnea); 3) a structured framework for cognitive evaluation including validated instruments (e.g., MoCA and AD8); 4) referrals to specialty care; 5) billing guidelines; and 6) community resources and materials that patients and care partners can use both pre‐ and post‐diagnosis. We have pilot‐tested these tools in UW primary care clinics and found high utilization both within and outside of targeted clinics indicating a halo effect of usage beyond those who attended the CME training. In this presentation, we will provide detailed descriptions of these Epic tools focusing on clinical application to total patient care. We plan to publish these workflow tools onto the Epic Digital Library to be downloaded and used by other health systems. Future plans include producing on‐demand CME training modules, a post‐diagnosis checklist, and applications for use of these tools in systems outside of Epic.

Fulfillment and Validity of the Kidney Health Evaluation Measure for People with Diabetes

ObjectiveTo evaluate the fulfillment and validity of the kidney health evaluation for people with diabetes (KED) Healthcare Effectiveness Data Information Set (HEDIS) measure. Patients and MethodsOptum Labs Data Warehouse (OLDW) was used to identify the nationally distributed US population aged 18 years and older, with diabetes, between January 1, 2017, and December 31, 2017. The OLDW includes deidentified medical, pharmacy, laboratory, and electronic health record (EHR) data. The KED fulfillment was defined in 2017 as both estimated glomerular filtration rate (eGFR) and urinary albumin-creatinine ratio testing within the measurement year. The KED validity was assessed using bivariate analyses of KED fulfillment with diabetes care measures in 2017 and chronic kidney disease (CKD) diagnosis and evidence-based kidney protective interventions in 2018. ResultsAmong eligible 5,635,619 Medicare fee-for-service beneficiaries, 736,875 Medicare advantage (MA) beneficiaries, and 660,987 commercial patients, KED fulfillment was 32.2%, 38.7%, and 37.7%, respectively. Albuminuria testing limited KED fulfillment with urinary albumin-creatinine ratio testing (<40%) and eGFR testing (>90%). The KED fulfillment was positively associated with receipt of diabetes care in 2017, CKD diagnosis in 2018, and evidence-based kidney protective interventions in 2018. The KED fulfillment trended lower for Black race, Medicare-Medicaid dual eligibility status, low neighborhood income, and low education status. ConclusionLess than 40% of adults with diabetes received guideline-recommended testing for CKD in 2017. Routine KED was associated with diabetes care and evidence-based CKD interventions. Increasing guideline-recommended testing for CKD among people with diabetes should lead to timely and equitable CKD detection and treatment.

A Knowledge-based and Data-driven Approach for Predicting Acute Kidney Injury in Patients with Heart Failure.

It has great potential to integrate medical knowledge and electronic health record data for diagnosis prediction. However, present studies only utilized information from knowledge graphs, omitting potentially significant global graph structural features. In this study, we proposed a knowledge and data integrating modeling approach to reconstruct patient electronic health record data with graph structure and use medical knowledge as internal information of patient data to build a risk prediction model for acute kidney injury in patients with heart failure based on graph neural networks. Experimental results based on the MIMIC III data showed that the method proposed was superior to other baseline models in predicting the risk of acute kidney injury in heart failure patients, with an accuracy of 0.725 and an F1 score of 0.755. This study provides a novel approach to the disease risk prediction models that integrates medical knowledge and data.

Multimodality Video Acquisition System for the Assessment of Vital Distress in Children.

In children, vital distress events, particularly respiratory, go unrecognized. To develop a standard model for automated assessment of vital distress in children, we aimed to construct a prospective high-quality video database for critically ill children in a pediatric intensive care unit (PICU) setting. The videos were acquired automatically through a secure web application with an application programming interface (API). The purpose of this article is to describe the data acquisition process from each PICU room to the research electronic database. Using an Azure Kinect DK and a Flir Lepton 3.5 LWIR attached to a Jetson Xavier NX board and the network architecture of our PICU, we have implemented an ongoing high-fidelity prospectively collected video database for research, monitoring, and diagnostic purposes. This infrastructure offers the opportunity to develop algorithms (including computational models) to quantify vital distress in order to evaluate vital distress events. More than 290 RGB, thermographic, and point cloud videos of each 30 s have been recorded in the database. Each recording is linked to the patient's numerical phenotype, i.e., the electronic medical health record and high-resolution medical database of our research center. The ultimate goal is to develop and validate algorithms to detect vital distress in real time, both for inpatient care and outpatient management.

A Survey on Medical Health Records and AI

As the population increases, so do the number of patients getting admitted in hospitals. This generates an overwhelming amount of data within the electronic health records (EHRs) that is impossible to manage manually. This is where machine learning concepts come in handy. The ML algorithms for regression-classification have become increasingly popular within the healthcare sector. Especially for cardiovascular diseases (CVD). Estimation states that by 2030, over 23 million people will die from CVD each year. But, it is estimated that 90% of CVD is preventable. The on time recognition and diagnosis of heart failure from the pre-existing medical records is a way to do so. However, the EHRs are not particularly reliable when it comes to the comparison of structured and unstructured data. This is due to the use of colloquial language and possible existence of sparse content. To tackle this issue, the proposed system uses the KTI-RNN model for recognition of unstructured data and removal of sparse content, TF-IWF model to extract the keyword set, the LDA model to extract the topic word set, and finally, the GA-BiRNN model to identify heart failure from extensive medical texts. The GA-BiRNN model is made up of a bidirectional recurrent neural network model and its output layer embedded with global attention mechanism and gating mechanism.

Does severity of periodontal disease influence the risk of developing gastrointestinal malignancies?

e15549 Background: In recent years, multiple observational studies have suggested increased risk of developing gastrointestinal (GI) malignancies in patients with prior periodontal disease (PD). Our aim was to determine whether a difference in incidence of GI malignancies exists amongst patients with pre-existing mild-moderate vs. severe PD. Methods: This was a single institution retrospective IRB approved study. We identified a cohort of patients between 2011-21 with PD and concurrent registration in independent dental and medical electronic health record systems at the University of Florida. Patients were stratified into 2 groups based on severity of PD (mild-moderate or severe). Descriptive statistics were used to compare the 2 groups based on demographics and medical history. Cumulative incidence curves were used to compare GI malignancy incidence between the 2 groups. Multivariable logistic regression was used to identify predictors of developing GI malignancies. Results: A total of 2168 patients with PD were identified with a median follow up of 4.7 (IQR 2.6-6.8) yr. Of these, 54% had severe PD and 46% had mild-moderate PD. Severe PD was more common within males (58% vs 50% in females), Black race (62% vs 52% in Caucasians), diabetics (59% vs 52% in non-diabetics), cigarette smokers (60% vs 51% in non-smokers) and excessive alcohol intake (70% vs 61% in moderate alcohol intake). There was no significant difference in cumulative incidence rates of GI malignancies (colon, pancreatic, gastric, hepatic, and esophageal) between the PD groups stratified by severity (12% in severe vs 11% in mild-moderate, p = 0.6 at 5 yr follow up). In univariable logistic regression, severe vs mild-moderate PD OR was 1.12 [95% CI: 0.87,1.46], thus more severe form of PD is not associated with developing a GI malignancy. In multivariable logistic regression, severe vs mild/moderate PD adjusted OR was 1.04 [95% CI: 0.79,1.36], thus also not statistically significant, with other risk factors of developing GI malignancies identified in Table 1. Conclusions: There was no association between the severity of periodontal disease and subsequent development of GI malignancy. [Table: see text]

Validation of diagnostic coding for bronchiectasis in an electronic health record system in Hong Kong.

Electronic medical record (EMR) databases can facilitate epidemiology research in various diseases including bronchiectasis. Given the diagnostic challenges of bronchiectasis, the validity of the coding in EMR requires clarification. We aimed to assess the validity of International Classification of Diseases, 9th Revision (ICD-9) code algorithms for identifying bronchiectasis in the territory-wide electronic medical health record system of Clinical Data Analysis and Reporting System (CDARS) in Hong Kong. Adult patients who had the diagnosis of bronchiectasis input from Queen Mary Hospital in 2011-2020 were identified using the ICD-9 code of 494 by CDARS. All patients who had high resolution computed tomography (HRCT) were reviewed by respiratory specialists to confirm the presence of bronchiectasis on HRCT. 19 617 patients who had the diagnostic code of bronchiectasis among all public hospitals in Hong Kong and 1866 in Queen Mary Hospital in the same period. 648 cases were randomly selected and validated using medical record and HRCT review by a respiratory specialist. The overall positive predictive value (PPV) was 92.7% (95% CI 90.7-94.7). This was the first ICD-9 coding validation for bronchiectasis in Hong Kong CDARS. Our study demonstrated that using ICD-9 code of 494 was reliable to support utility of CDARS database for further clinical research on bronchiectasis. This article is protected by copyright. All rights reserved.

Nursing protocol for heart failure remote monitoring in patients with ICD/CRT with a 5 variable multiparametric tool

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Many patients with heart failure (HF) suffer from decompensation that sometimes leads to hospitalizations, harming prognosis and quality of life. Some cardiac implantable electronic devices (CIEDs) allow the transmission of clinical data online, providing helpful information without the patient having to go to a face-to-face consultation, which usually takes no less than 25 minutes. Moreover, a 5-variable multiparametric tool (S1 and S3 heart sounds, pulmonary congestion through a thoracic impedance, volume and respiratory rate, night heart rate) expresses a heart failure index (HM-5) incorporated into some CIEDs, capable of predicting 34 days before HF decompensation episodes when the reported index is ≥16, with a sensitivity of 70%. According to this warning, transmissions with a potential risk of HF exacerbation should be classified. Ideally, reviewing these warnings should not take more time than is invested in the face-to-face follow-up. Purpose Create a protocol with a risk classification according to the heart failure index and compare the duration of face-to-face and remote online follow-up performed by the same nursing staff in patients with CIEDs with a heart failure index activated in the arrhythmia consultation in a tertiary hospital. Methods First, we created a colour scale protocol (red alert: urgent event, yellow alert: important event and green alert: normal state) classifying heart failure index alerts during one year (January – December 2021) (See figure 1). Patient clinical characteristics, and duration of face-to-face &amp; remote monitoring follow-ups were registered, quantifying the time spent reviewing each patient. Results Twenty-two patients with CIEDs and HM-5 tool activated were included; 28% were female, 72% had hypertension, 40,9% had diabetes mellitus, 22,7% had chronic kidney disease, and the mean left ventricle ejection fraction was 36,18%, half of them has ischaemic cardiomyopathy. The annexed protocol was followed with the help of the medical health record platform that classified the transmissions according to a colour scale. The rate of alerts/patient-year was 0.97; 22% of yellow alerts were related to HF. The mean time nursing review face-to-face versus remote was 16.55 min vs 3.05 min (p&amp;lt;0.001) (table 1). Conclusion Remote follow-up by the nursing staff of patients with HF and ICD provided with a 5-variable multiparametric tool is strict and protocolized in our hospital. It requires less time compared to face-to-face review.

NFT-based online prescription for data management in healthcare services

ABSTRACT The development of electronic health (e-health) systems in the last decades led to the emergence of effective infrastructures for managing health records. Due to the vitality and massive amount of healthcare records, blockchain is considered one of the most important infrastructures to store and exchange health records. Access to patient’s medical histories and health record management are essential to healthcare systems. Patients’ records must be exchanged among various parts of the healthcare system and users should have access to manage their profiles and use online facilities in a secure manner. In this paper, for the first time, the non-fungible token (NFT), which is a nascent technology, is used to certify the information immutability of healthcare records in the blockchain infrastructure. The ownership of property and prevention of information disclosure is guaranteed using the encryption strategy is used in the proposed NFT technology. Furthermore, the proposed model satisfies all security criteria such as data integrity, non-repudiation, anonymity, confidentially and privacy. As a result, the proposed model outperforms the related research in the healthcare domain. The proposed model brings many facilities for both patients and physicians such as secure accessibility, data quality, patient consent, efficient record management and better availability and reliability.

Medical Health Record using Block Chain

The future of healthcare systems is being shaped by incorporating emerged technological innovations to drive new models for patient care. By acquiring, integrating, analyzing, and exchanging medical data at different system levels, new practices can be introduced, offering a radical improvement to healthcare services. We present a novel smart and secure Medical health record system which, leverages blockchain technologies, permits patient centric medical records transactions .Office of the National Coordinator (ONC) for Health Information Technology is seeking patient-centric MHR designs that shift data ownership from providers to patients. There are multiple barriers to patient-centric MHR in the current system, such as security and privacy concerns, data inconsistency, timely access to the right records across multiple healthcare facilities. After investigating the current workflow of MHR, our system provides a feasible solution to these challenges by utilizing the unique features of blockchain. In particular, we develop a blockchain-based architecture and enable a flexible configuration thereof, which optimize medical data sharing between different health entities and fulfill the diverse levels of Quality of Service (QoS) that MHR may require. Finally, we highlight the benefits of the proposed MHR system and possible directions for future research.

A survey: medical health record data security based on interplanetary file system and blockchain technologies

The adoption of modern health records is growing more mature, yet security issues always accompany it. Interplanetary file system (IPFS) and blockchain are developing technologies with decentralization, distributed fault tolerance, and trustworthiness. Using IPFS and blockchain technology to tackle medical health record data security issues is a very promising trend, and it is presently being utilized to secure medical health record data security. This article first explains the idea of IPFS and highlights the classification of existing IPFS and blockchain techniques before briefly discussing distributed ledger to tackle the existing medical health record data security challenges and faults. Finally, to preserve medical health records, a new medical health record storage architectural model based on IPFS and blockchain technologies is presented.

A Structured Analysis to study the Role of Machine Learning and Deep Learning in The Healthcare Sector with Big Data Analytics.

Machine and deep learning are used worldwide. Machine Learning (ML) and Deep Learning (DL) are playing an increasingly important role in the healthcare sector, particularly when combined with big data analytics. Some of the ways that ML and DL are being used in healthcare includePredictive Analytics,Medical Image Analysis,Drug Discovery,Personalized Medicine,and Electronic Health Records (EHR) Analysis.It has become one of the advanced and popular tool forcomputer sciencedomain.' Theadvancementof ML and DL forvariousfieldshas opened new avenues for research and development. It could revolutionize prediction and decision-makingcapabilities. Due to increased awarenessabout the ML and DL in the healthcare, it has become one of the vital approaches for thesector. High-volumeof unstructured, and complex medical imaging data from health monitoring devices, gadgets, sensors, etc. Is the biggest trouble forhealthcare sector. The current study uses analysis to examine research trends inadoption of machine learning and deep learningapproaches in the healthcare sector. The WoS database for SCI/SCI-E/ESCIjournals areused as the datasetsfor the comprehensive analysis. Apart from these various search strategy are utilised for therequisitescientificanalysis of the extracted research documents. Bibliometrics R statistical analysis is performed for year-wise, nation-wise, affiliation-wise, research area, sources, documents, and authorbased analysis. VOS viewer software is used to create author, source, country, institution, global cooperation, citation, co-citation, and trending term co-occurrence networks. ML and DL, combined with big data analytics, have the potential to revolutionize healthcare by improving patient outcomes, reducing costs, and accelerating the development of new treatments, so the currentstudy will help academics, researchers, decision-makers, and healthcare professionals understand and direct research.

Validation of Diagnostic Coding for Asthma in an Electronic Health Record System in Hong Kong.

Electronic health record (EHR) databases can facilitate epidemiology research into various diseases including asthma. Given the diagnostic challenges of asthma, the validity of the coding in EHR requires clarification. We aimed to assess the validity of International Classification of Diseases, 9th Revision (ICD-9) code algorithms for identifying asthma in the territory-wide electronic medical health record system of the Clinical Data Analysis and Reporting System (CDARS) in Hong Kong. Adult patients who had the diagnosis of asthma input from all public hospitals in Hong Kong and those from Queen Mary Hospital in 2011-2020 were identified using the ICD-9 code of 493 (493.0, 493.1, 493.2, and 493.9) by CDARS. Patients' clinical record and spirometry were reviewed by two respiratory specialists to confirm the presence of asthma in the randomly selected cases. There were 43,454 patients who had the diagnostic code of asthma among all public hospitals in Hong Kong and 1852 in Queen Mary Hospital in the same period. A total of 200 cases were randomly selected and validated using medical record and spirometry review by a respiratory specialist. The overall positive predictive value (PPV) was 85.0% (95% CI 80.1-89.9%). This was the first ICD-9 code validation for CDARS (EHR) in Hong Kong on asthma. Our study demonstrated that using ICD-9 code (493.0, 493.1, 493.2 and 493.9) to identify asthma can result in a PPV that was reliable to support the utility of the CDARS database for further research on asthma among the Hong Kong population.