Accuracy of machine learning in the diagnosis of odontogenic cysts and tumors: a systematic review and meta-analysis.

BACKGROUND: The recent impact of artificial intelligence in diagnostic services has been enormous. Machine learning tools offer an innovative alternative to diagnose cysts and tumors radiographically that pose certain challenges due to the near similar presentation, anatomical variations, and superimposition. It is crucial that the performance of these models is evaluated for their clinical applicability in diagnosing cysts and tumors. METHODS: A comprehensive literature search was carried out on eminent databases for published studies between January 2015 and December 2022. Studies utilizing machine learning models in the diagnosis of odontogenic cysts or tumors using Orthopantomograms (OPG) or Cone Beam Computed Tomographic images (CBCT) were included. QUADAS-2 tool was used for the assessment of the risk of bias and applicability concerns. Meta-analysis was performed for studies reporting sufficient performance metrics, separately for OPG and CBCT. RESULTS: 16 studies were included for qualitative synthesis including a total of 10,872 odontogenic cysts and tumors. The sensitivity and specificity of machine learning in diagnosing cysts and tumors through OPG were 0.83 (95% CI 0.81-0.85) and 0.82 (95% CI 0.81-0.83) respectively. Studies utilizing CBCT noted a sensitivity of 0.88 (95% CI 0.87-0.88) and specificity of 0.88 (95% CI 0.87-0.89). Highest classification accuracy was 100%, noted for Support Vector Machine classifier. CONCLUSION: The results from the present review favoured machine learning models to be used as a clinical adjunct in the radiographic diagnosis of odontogenic cysts and tumors, provided they undergo robust training with a huge dataset. However, the arduous process, investment, and certain ethical concerns associated with the total dependence on technology must be taken into account. Standardized reporting of outcomes for diagnostic studies utilizing machine learning methods is recommended to ensure homogeneity in assessment criteria, facilitate comparison between different studies, and promote transparency in research findings.

Surveillance of stroke: a South-East Asia Region (SEAR) perspective.

Surveillance of stroke is critical to track its burden and assess progress in prevention and treatment. We reviewed the literature to evaluate stroke surveillance efforts in the South-East Asia Region (SEAR) countries, identify progress and assess gaps. Epidemiological data on all the major parameters such as the incidence, prevalence and mortality of stroke were available for India and Thailand but for none of the other SEAR countries. Most of the epidemiological data came from investigator-initiated studies. National stroke surveillance was present only in India in the form of a National Stroke Registry Programme and Thailand has a national database that was used to obtain epidemiological data for stroke. Research on novel methods for stroke registration, such as using information technology, was absent. This review identified serious gaps in the monitoring and surveillance of stroke in SEAR countries. Systematic efforts are needed to fill those gaps.

Stroke systems of care in South-East Asia Region (SEAR): commonalities and diversities.

The Southeast Asia Region (SEAR) accounts for nearly 50% of the developing world's stroke burden. With various commonalities across its countries concerning health services, user awareness, and healthcare-seeking behavior, SEAR still presents profound diversities in stroke-related services across the continuum of care. This review highlights the numerous systems and challenges in access to stroke care, acute stroke care services, and health care systems, including rehabilitation. The paper has also attempted to compile information on the availability of stroke specialized centers, Intravenous thrombolysis (IVT) ready centers, Endovascular therapy (EVT) ready centers, rehabilitation centers, and workforce against a backdrop of each country's population. Lastly, the efforts of WHO (SEARO)-CMCL (World Health Organization-South East Asia region, Christian Medical College & Hospital Ludhiana) collaboration towards improving stroke services and capacity among the SEAR have been described.

The burden, risk factors and unique etiologies of stroke in South-East Asia Region (SEAR).

The World Health Organization (WHO) South East Asia Region (SEAR) comprises 11 countries, which are one of the most culturally, topographically, and socially diverse areas worldwide, undergoing an epidemiological transition towards non-communicable diseases, including stroke and other cardiovascular diseases (CVDs). This region accounts for over 40% of the global stroke mortality. Few well-designed population-based epidemiological studies on stroke are available from SEAR countries, with considerable variations among them. Ischemic stroke, a common stroke subtype, has higher frequencies of intracerebral hemorrhage in many countries. Along with an aging population, the increased prevalence of risk factors such as hypertension, diabetes mellitus, tobacco and alcohol consumption, lack of physical activity, high ambient pollution, heat, and humidity contribute to the high burden of stroke in this region. SEAR's many unique and uncommon stroke etiologies include cerebral venous thrombosis, tuberculosis, dengue, scrub typhus, falciparum malaria, snake bite, scorpion sting, etc. Current data on stroke burden and risk factors is lacking, compelling an urgent need for high-quality hospital-level and population-level data in all SEAR countries. Strategies towards a consolidated approach for implementing improved stroke prevention measures, stroke surveillance, and established stroke systems of care are the path to bridging the gaps in stroke care.

Healthcare professionals' perspectives of the provision of, and challenges for, eating, drinking and psychological support post stroke: findings from semistructured interviews across India.

AIM: This qualitative study explores with health professionals the provision of, and challenges for, postdischarge stroke care, focussing on eating, drinking and psychological support across India. DESIGN: Qualitative semistructured interviews. SETTING: Seven geographically diverse hospitals taking part in a Global Health Research Programme on Improving Stroke Care in India. PARTICIPANTS: A purposive sample of healthcare professionals with current experience of working with patients who had a stroke. RESULTS: Interviews with 66 healthcare professionals (23 nurses (14 staff nurses; 7 senior nurse officers; 1 intensive care unit nurse; 1 palliative care nurse)); 16 doctors (10 neurologists; 6 physicians); 10 physiotherapists; 5 speech and language therapists; 4 occupational therapists; 4 dieticians; 2 psychiatrists; and 2 social workers resulted in three main themes: integrated inpatient discharge care planning processes; postdischarge patient and caregiver role and challenges; patient and caregiver engagement post discharge. CONCLUSIONS: Discharge planning was integrated and customised, although resources were limited in some sites. Task shifting compensated for a lack of specialists but was limited by staff education and training. Caregivers faced challenges in accessing and providing postdischarge care. Postdischarge care was mainly hospital based, supported by teleservices, especially for rural populations. Further research is needed to understand postdischarge care provision and the needs of stroke survivors and their caregivers.

Dental implants in growing patients: A quality assessment of systematic reviews.

Background: Dental implants in the recent past have become indispensable in restoring loss of space and support as well as aiding the aesthetics. Although in practice for a few years there is no consensus on the protocols for placement and usage of dental implants in growing jaws. Objective: The present study aims to evaluate evidence-based literature on single or multiple dental implant placements in children(up to 17 years) and to identify areas lacking and gaps in knowledge. The quality assessment tool,namely AMSTAR-2, will also aim to evaluate the quality of said research around paediatric dental implants. Design: The study was prospectively registered on the Open Science Framework https://osf.io/e59bt/?view_only=ec8fb69455c240ecbfc7379734784bf7.For source selection, electronic searches were performed on MEDLINE, EMBASE, Scopus databases as well as Google Scholar for all English language systematic reviews and meta analysis on dental implants placed on children up to 17yrs of age by 2 reviewers, wherein the publications until December 2020 were included. A final dataset of 4 systematic reviews were incorporated and analysed using the AMSTAR-2 grading tool. Results: It was seen that only one study showed moderate overall confidence while one low overall confidence as per the AMSTAR-2. The other 2 systematic reviews were of critically low confidence levels. Hence, the existing systematic reviews on the subject are not of high quality. Most of the reviews have no consensus regarding use and placement of implants in growing patients. Conclusion: More research and stricter adherence to the quality assessment guidelines is recommended for all future systematic reviews regarding dental implant in children.

Secondary prevention by structured semi-interactive stroke prevention package in INDIA (SPRINT INDIA): Findings from the process evaluation of a randomized controlled trial.

INTRODUCTION: Secondary Prevention by Structured Semi-Interactive Stroke Prevention Package in INDIA Trial delivered secondary stroke awareness intervention to sub-acute stroke patients in form of workbook, videos and SMS across 31 centres in 12 languages. Trial was stopped for futility due to fewer vascular outcomes than anticipated. Trial results indicated that trial intervention, did not lead to reduction in vascular events. We carried out process evaluation, to evaluate trial implementation and participant's perspectives, to comprehend the trial's futile outcomes. MATERIALS AND METHODS: Using mixed methods approach, qualitative interviews and quantitative data from case report forms, workbooks and questionnaires were analysed to measure intervention fidelity and contamination. Using purposive sampling, 115 interviews of patient-caregiver dyads and health professionals at 11 centres and 2 focus group discussions were held. RESULTS AND DISCUSSION: Iterative thematic analysis of qualitative data was done with RE-AIM and realist models. There was good fidelity to intervention and adherence to protocol; however, there was dilution of inclusion criteria by randomly enrolling uneducated and caregiver-dependent patients. Centre coordinators provided counselling to both arms, not specified by protocol, causing bias. Coordinators found it difficult to keep patients motivated to view intervention which was corroborated by fidelity questionnaire showing decreased viewing of intervention for a year. Cardiovascular protection improved in routine care by virtue of participating in trial. No contamination of intervention was reported. CONCLUSION: The intervention was acceptable by patients and caregivers, which could be made a community-based programme. Reasons identified for decreased viewing were repetitive content and non-availability of personal cellular device.