This article attempts to investigate the various types of threats that exist in healthcare information systems (HIS). A study has been carried out in one of the government-supported hospitals in Malaysia.The hospital has been equipped with a Total Hospital Information System (THIS). The data collected were from three different departments, namely the Information Technology Department (ITD), the Medical Record Department (MRD), and the X-Ray Department, using in-depth structured interviews. The study identified 22 types of threats according to major threat categories based on ISO/IEC 27002 (ISO 27799:2008). The results show that the most critical threat for the THIS is power failure followed by acts of human error or failure and other technological factors. This research holds significant value in terms of providing a complete taxonomy of threat categories in HIS and also an important component in the risk analysis stage.
This study aims to investigate healthcare practitioner behaviour in adopting Health Information Systems which could affect patients' safety and quality of health. A qualitative study was conducted based on a semi-structured interview protocol on 31 medical doctors in three Malaysian government hospitals implementing the Total Hospital Information Systems. The period of study was between March and May 2015. A thematic qualitative analysis was performed on the resultant data to categorize them into relevant themes. Four themes emerged as healthcare practitioners' behaviours that influence the unsafe use of Hospital Information Systems. The themes include (1) carelessness, (2) workarounds, (3) noncompliance to procedure, and (4) copy and paste habit. By addressing these behaviours, the hospital management could further improve patient safety and the quality of patient care.
Advancements in electronic health record system allow patients to store and selectively share their medical records as needed with doctors. However, privacy concerns represent one of the major threats facing the electronic health record system. For instance, a cybercriminal may use a brute-force attack to authenticate into a patient's account to steal the patient's personal, medical or genetic details. This threat is amplified given that an individual's genetic content is connected to their family, thus leading to security risks for their family members as well. Several cases of patient's data theft have been reported where cybercriminals authenticated into the patient's account, stole the patient's medical data and assumed the identity of the patients. In some cases, the stolen data were used to access the patient's accounts on other platforms and in other cases, to make fraudulent health insurance claims. Several measures have been suggested to address the security issues in electronic health record systems. Nevertheless, we emphasize that current measures proffer security in the short-term. This work studies the feasibility of using a decoy-based system named HoneyDetails in the security of the electronic health record system. HoneyDetails will serve fictitious medical data to the adversary during his hacking attempt to steal the patient's data. However, the adversary will remain oblivious to the deceit due to the realistic structure of the data. Our findings indicate that the proposed system may serve as a potential measure for safeguarding against patient's information theft.
Current suicide risk assessments for predicting suicide attempts are time consuming, of low predictive value and have inadequate reliability. This paper aims to develop a predictive model for suicide attempts among patients with depression using machine learning algorithms as well as presents a comparative study on single predictive models with ensemble predictive models for differentiating depressed patients with suicide attempts from non-suicide attempters. We applied and trained eight different machine learning algorithms using a dataset that consists of 75 patients diagnosed with a depressive disorder. A recursive feature elimination was used to reduce the features via three-fold cross validation. An ensemble predictive models outperformed the single predictive models. Voting and bagging revealed the highest accuracy of 92% compared to other machine learning algorithms. Our findings indicate that history of suicide attempt, religion, race, suicide ideation and severity of clinical depression are useful factors for prediction of suicide attempts.
A CDSS generates a high number of inappropriate alerts that interrupt the clinical workflow. As a result, clinicians silence, disable, or ignore alerts, thereby undermining patient safety. Therefore, the effectiveness and appropriateness of CDSS alerts need to be evaluated. A systematic review was carried out to identify the factors that affect CDSS alert appropriateness in supporting clinical workflow. Seven electronic databases (PubMed, Scopus, ACM, Science Direct, IEEE, Ovid Medline, and Ebscohost) were searched for English language articles published between 1997 and 2018. Seventy six papers met the inclusion criteria, of which 26, 24, 15, and 11 papers are retrospective cohort, qualitative, quantitative, and mixed-method studies, respectively. The review highlights various factors influencing the appropriateness and efficiencies of CDSS alerts. These factors are categorized into technology, human, organization, and process aspects using a combination of approaches, including socio-technical framework, five rights of CDSS, and Lean. Most CDSS alerts were not properly designed based on human factor methods and principles, explaining high alert overrides in clinical practices. The identified factors and recommendations from the review may offer valuable insights into how CDSS alerts can be designed appropriately to support clinical workflow.
The nursing schedule generation is an important activity that takes a considerable amount of time for managers to prepare and amend. It involves the optimal allocation of nurses to shifts, factoring various constraints like shift timings, holidays, leaves, and emergencies. This paper provides the design and development details for an automated nurse scheduling system called "ROTA," implemented for a 2032 bed multi-specialty tertiary teaching hospital, having 1800 staff nurses and 98 wards. The system generates daily, weekly, monthly schedules, nurse face sheets, duty allocation charts, swapping schedules, and training details for nurses. The system improved managerial control and saved a considerable amount of time for nurses to prepare the schedule. A survey conducted to gauge the system's satisfaction level showed that 91% of nurses were satisfied with ROTA. Overall, the system saved 78% of nurse scheduling time, resulting in a 3% cost reduction for the hospital.
This study modelled factors that predict fake news sharing during the COVID-19 health crisis using the perspective of the affordance and cognitive load theory. Data were drawn from 385 social media users in Nigeria, and Partial Least Squares (PLS) was used to analyse the data. We found that news-find-me perception, information overload, trust in online information, status seeking, self-expression and information sharing predicted fake news sharing related to COVID-19 pandemic among social media users in Nigeria. Greater effects of news-find-me perception and information overload were found on fake news sharing behaviour as compared to trust in online information, status seeking, self-expression and information sharing. Theoretically, our study enriches the current literature by focusing on the affordances of social media and the abundance of online information in predicting fake news sharing behaviour among social media users, especially in Nigeria. Practically, we suggest intervention strategies which nudge people to be sceptical of the information they come across on social media.
This study aimed to evaluate the effect of a novel progressive web application (PWA) on the patient's oral and denture knowledge and hygiene. Fifty-two removable partial denture wearers were randomised to receive education using the PWA, or verbal instructions accompanied by demonstration of hygienic procedures. Changes in the participants' knowledge score, plaque index, gingival index and denture plaque was evaluated during a follow-up period of 3 months. The participants' acceptance of PWA was explored through usage logs and a feedback form. Both groups showed significant improvement in knowledge scores, oral and denture hygiene indices (p < 0.001) after education. The PWA group demonstrated significantly lower gingival index score than control (p = 0.008) at the third month review. In conclusion, there is potential of using mobile application in educating elderly patients and the PWA is a viable option for providing post-denture delivery instructions.
As a chronic disease, diabetes mellitus has emerged as a worldwide epidemic. The aim of this study is to classify diabetes disease by developing an intelligence system using machine learning techniques. Our method is developed through clustering, noise removal and classification approaches. Accordingly, we use expectation maximization, principal component analysis and support vector machine for clustering, noise removal and classification tasks, respectively. We also develop the proposed method for incremental situation by applying the incremental principal component analysis and incremental support vector machine for incremental learning of data. Experimental results on Pima Indian Diabetes dataset show that proposed method remarkably improves the accuracy of prediction and reduces computation time in relation to the non-incremental approaches. The hybrid intelligent system can assist medical practitioners in the healthcare practice as a decision support system.
The study investigated the effect of different input selections on the performance of artificial neural networks in screening for acute myocardial infarction (AMI) in Malaysian patients complaining of chest pain. We used hospital data to create neural networks with four input selections and used these to diagnose AMI. A 10-fold cross-validation and committee approach was used. All the neural networks using various input selections outperformed a multiple logistic regression model, although the difference was not statistically significant. The neural networks achieved an area under the ROC curve of 0.792 using nine inputs, whereas multiple logistic regression achieved 0.739 using 64 inputs. Sensitivity levels of over 90 per cent were achieved using low output threshold levels. Specificity levels of over 90 per cent were achieved using threshold levels of 0.4-0.5. Thus neural networks can perform as well as multiple logistic regression models even when using far fewer inputs.
Effective delivery of post-insertion instructions is essential for denture care and oral health. This study aimed to develop a progressive web application (PWA) to educate patients' chairside and serve as a reference material. A need analysis was conducted before prototype development. Subsequently, the prototype was subjected to content verification, design appraisal and usability testing. The results of usability testing revealed a user task success rate of 94.4%, with an adjusted Wald 95% confidence interval of 83-100%. User satisfaction assessed using the Single Ease Questionnaire and System Usability Score reported a mean score of 6.13 (95% CI: 5.69-6.55) and 85.9 (95% CI: 82.2-89.6), respectively, indicating good usability. This study highlights the systematic approach of developing an evidence-based educational PWA to meet the usability standards for mobile applications. This PWA is useful in clinical studies to explore mobile technologies' potential in educating denture wearers, especially in the older population.
The objective of this study is to identify factors influencing unsafe use of hospital information systems in Malaysian government hospitals. Semi-structured interviews with 31 medical doctors in three Malaysian government hospitals implementing total hospital information systems were conducted between March and May 2015. A thematic qualitative analysis was performed on the resultant data to deduce the relevant themes. A total of five themes emerged as the factors influencing unsafe use of a hospital information system: (1) knowledge, (2) system quality, (3) task stressor, (4) organization resources, and (5) teamwork. These qualitative findings highlight that factors influencing unsafe use of a hospital information system originate from multidimensional sociotechnical aspects. Unsafe use of a hospital information system could possibly lead to the incidence of errors and thus raises safety risks to the patients. Hence, multiple interventions (e.g. technology systems and teamwork) are required in shaping high-quality hospital information system use.
Radiology reporting is narrative, and its content depends on the clinician's ability to interpret the images accurately. A tertiary hospital, such as anonymous institute, focuses on writing reports narratively as part of training for medical personnel. Nevertheless, free-text reports make it inconvenient to extract information for clinical audits and data mining. Therefore, we aim to convert unstructured breast radiology reports into structured formats using natural language processing (NLP) algorithm. This study used 327 de-identified breast radiology reports from the anonymous institute. The radiologist identified the significant data elements to be extracted. Our NLP algorithm achieved 97% and 94.9% accuracy in training and testing data, respectively. Henceforth, the structured information was used to build the predictive model for predicting the value of the BIRADS category. The model based on random forest generated the highest accuracy of 92%. Our study not only fulfilled the demands of clinicians by enhancing communication between medical personnel, but it also demonstrated the usefulness of mineable structured data in yielding significant insights.
The paediatric orthopaedic expert system analyses and predicts the healing time of limb fractures in children using machine learning. As far we know, no published research on the paediatric orthopaedic expert system that predicts paediatric fracture healing time using machine learning has been published. The University Malaya Medical Centre (UMMC) offers paediatric orthopaedic data, comprises children under the age of 12 radiographs limb fractures with ages recorded from the date and time of initial trauma. SVR algorithms are used to predict and discover variables associated with fracture healing time. This study developed an expert system capable of predicting healing time, which can assist general practitioners and healthcare practitioners during treatment and follow-up. The system is available online at https://kidsfractureexpert.com/.