Clinical Algorithm Research Articles

Continuous cardiac monitoring in epilepsy: an implantable loop manual activation algorithm for improving ECG signal acquisition accuracy

BackgroundThe muscle artifacts, caused by prominent muscle contractions, mimicking cardiac arrhythmias, might compromise the ECG signal quality and the implantable loop recorder memory capacity in patients with epilepsy. We developed an epileptic seizures clinical pattern-based implantable loop recorder manual activation algorithm, presenting its real-world efficacy here.MethodsOne hundred ninety-three patients (18–60 years) with drug-resistant focal epilepsy were consecutively enrolled and underwent a subcutaneous loop recorder implantation. Patients with focal onset-aware seizures and patients with focal impaired awareness seizures /bilateral tonic-clonic seizures without aura were recommended to use the activator once - just after the episode. Patients with focal impaired awareness seizures/bilateral tonic-clonic seizures with aura, the caregivers of patients experiencing status epilepticus, were advised to use the activator twice - during the aura and after the episode/ regaining consciousness.ResultsSix thousand four hundred ninety-four ECG traces (4826 - auto-triggered events, 1668 - person-activated events) were recorded and analyzed. The rate of true positive events in the person-activated group was statistically higher than in the autoactivation group (72.5% vs.19.4%, p < 0.0001). Person-activated false-positive events were observed in 30.5% of patients with focal impaired awareness seizures and 27.7% in patients with bilateral tonic-clonic seizures. The highest rate of false-positive events (61.5%) was detected in patients undergoing epileptic status, and the lowest rate (3.8%) - was in patients with focal onset aware seizures. The rate of false-positive events was significantly higher in patients with impaired awareness seizures without aura both in focal impaired awareness (45.5% vs. 19.3%, p < 0.0001) and bilateral tonic-clonic seizure groups (38.8% vs. 5.9%, p < 0.0001).ConclusionsArrhythmias with varying clinical outcomes are expected in epilepsy patients and have been monitored continuously. The specified loop recorder external activation algorithm can improve the clinically relevant cardiac arrhythmia detection accuracy in epilepsy patients and the value of future studies.

Deintensification of potentially inappropriate medications amongst older frail people with type 2 diabetes: Protocol for a cluster randomised controlled trial (D-MED study)

AimsAmongst elderly people with type 2 diabetes (T2D) over prescribing can result in emergency ambulance call-outs, falls and fractures and increased mortality, particularly in frail patients. Current clinical guidelines, however, remain focused on medication intensification rather than deintensification where appropriate. This study aims to evaluate the effectiveness of an electronic decision-support system and training for the deintensification of potentially inappropriate medications amongst older frail people with T2D, when compared to ‘usual’ care at 12-months. MethodsThis study is an open-label, multi-site, two-armed pragmatic cluster-randomised trial. GP practices randomised to the ‘enhanced care’ group have an electronic decision support system installed and receive training on the tool and de-intensification of diabetes medications. The system flags eligible patients for possible deintensification of diabetes medications, linking the health care professional to a clinical algorithm. The primary outcome will be the number of patients at 12-months who have had potentially inappropriate diabetes medications de-intensified. ResultsStudy recruitment commenced in June 2022. Data collection commenced in January 2023. Baseline data have been extracted from 40 practices (3145 patients). ConclusionsDigital technology, involving computer decision systems, may have the potential to reduce inappropriate medications and aid the process of de-intensification. Trial registrationInternational Standard Randomised Controlled Trial Number: ISRCTN53221378. Available at: https://www.isrctn.com/ISRCTN53221378.

Clinical and genetic diagnosis of familial hypercholesterolaemia in patients undergoing coronary angiography: the Ludwigshafen Risk and Cardiovascular Health Study.

To investigate the prevalence of familial hypercholesterolaemia (FH) and compare the performance of clinical criteria and genetic testing in patients undergoing coronary angiography. The prevalence of FH was determined with the Dutch Lipid Clinical Network (DLCN), US-MEDPED, Simon Broome (SB) criteria, the 'Familial Hypercholesterolaemia Case Ascertainment Tool' (FAMCAT), and a clinical algorithm. Genetic screening was conducted with a custom array from Affymetrix (CARRENAL array) harboring 944 FH mutations. The study cohort consisted of 3267 patients (78.6% with coronary artery disease [CAD]). FH was diagnosed in 2.8%, 2.2%, 3.9%, and 7.9% using the DLCN, US-MEDPED, SB criteria, and the FAMCAT. The clinical algorithm identified the same patients as the SB criteria. Pathogenic FH mutations were found in 1.2% (1.2% in patients with CAD, 1.0% in patients without CAD). FH was more frequently diagnosed in younger patients. With genetic testing as reference, the clinical criteria achieved areas under the ROC curve (AUCs) in the range of 0.56-0.68. Using only LDL-C corrected for statin intake, an AUC of 0.68 was achieved. FH is up to fourfold more prevalent in patients undergoing coronary angiography than in contemporary cohorts representing the general population. Different clinical criteria yield substantially different diagnosis rates, overestimating the prevalence of FH compared to genetic testing. LDL-C testing alone may be sufficient to raise the suspicion of FH, which then needs to be corroborated by genetic testing.

Prognostic Impact of Preoperative Thrombocytosis on Recurrence-Free Survival in Patients with Upper Tract Urothelial Carcinoma.

The aim of this work was to evaluate the prognostic potential of preoperative thrombocytosis for recurrence-free survival (RFS) and cancer-specific survival (CSS) among patients subjected to radical nephroureterectomy (RNU) due to UTUC. Analytical cohort was composed of a single-center series of 405 patients treated between January 1999 and December 2020. Thrombocytosis was defined as a platelet count exceeding the threshold value of 400 × 109 per L. Along with the Kaplan-Meier survival probability, Cox proportional hazard regression models were used. Preoperative thrombocytosis confirmed in 71 patients (17.5%) was significantly associated with the higher pathological tumor stage, lymph node metastasis, prior bladder cancer diagnosis, and preoperative anemia. With a median post-surgical follow-up period of 33.5 months, 125 patients (30.9%) experienced disease recurrence. The recurrence rate among patients with normal platelet levels was 13.6%, compared with 22.2% in those with preoperative thrombocytosis (p < 0.03). The 5-year RFS estimates reached 36.6% in the thrombocytosis-confirmed group. Multivariate analysis implied that preoperative thrombocytosis was a significant independent prognosticator of both poor RFS (HR 2.22, 95% CI 1.14-4.31, p = 0.02) and CSS (HR 2.48, 95% CI 1.14-3.09, p = 0.01). Patients with a clinically significant elevation of platelet count prior to RNU were more likely to have UTUC with advanced tumor stages and lymph node metastases. Preoperative thrombocytosis was an independent predictor of RFS and CSS in patients who underwent radical nephroureterectomy. Furthermore, preoperative thrombocytosis may complement and refine UTUC clinical prediction algorithms as an independent indicator of adverse survival outcomes.

A clinician's guide to the 2023 VA/DoD Clinical Practice Guideline for Management of Posttraumatic Stress Disorder and Acute Stress Disorder.

A clinical practice guideline (CPG) is a rigorously established set of recommendations based on currently available evidence about the efficacy, safety, acceptability, and feasibility of interventions to assist with clinical decision-making. The 2023 Department of Veterans Affairs /Department of Defense Clinical Practice Guideline for Management of Posttraumatic Stress Disorder and Acute Stress Disorder is described herein. The CPG recommendations are accompanied by a clinical algorithm, which incorporates principles of evidence-based practice, shared decision-making, and functional and contextual assessments of goals and outcomes. An overview of the CPG recommendations is combined with a discussion of questions that clinicians and patients may face in implementing the CPG and suggestions for how to effectively work with the CPG.

Clinical algorithm model based on cfDNA to predict SLE disease activity.

Circulating cell-free DNA (cfDNA) has been widely used as a new liquid-biopsy marker. Dysregulation of cfDNA has been found in patients with systemic lupus erythematosus (SLE). However, the detailed association between cfDNA and SLE has not been thoroughly studied. Plasma samples were collected from 88 patients with active SLE and 39 patients with inactive SLE. The cfDNA concentration was determined, and the length and distribution of cfDNA fragments were verified. cfDNA concentrations were significantly higher in patients with active SLE than in patients with inactive SLE (0.4 [0.18-0.897] ng/µL vs 0.249 [0.144-0.431] ng/µL; p = .043). cfDNA fragments were enriched in the ranges of 153-198 bp and 300-599 bp. cfDNA concentrations were associated with the reduction of the anti-double-stranded DNA (dsDNA) antibodies titer (r = -0.301, p = .034). The presence of anti-U1 ribonucleoprotein (p = .012), anti-Sjogren syndrome A (p = .024), anti-dsDNA (p = .0208), and anti-nucleosome antibodies (p = .0382) might associate to the variation of cfDNA concentration. Reduced cfDNA concentration was associated with renal damage in active SLE patients (0.31 [0.11-0.73] ng/µL vs 0.65 [0.27-1.53] ng/µL; p = .009). The Active index, a combination model including cfDNA concentration and other clinical indices, had an area of 0.886 under the receiver operating characteristics curve for distinguishing active SLE. The Active index was positively correlated with the SLE disease activity index score (r = 0.6724, p < .0001). Through systematic stratified analysis and clinical algorithm model, this study found that plasma cfDNA concentration is closely related to SLE disease severity, which has guiding significance for the future clinical application of cfDNA in SLE.

Unveiling the future of COVID-19 patient care: groundbreaking prediction models for severe outcomes or mortality in hospitalized cases.

Previous studies have identified COVID-19 risk factors, such as age and chronic health conditions, linked to severe outcomes and mortality. However, accurately predicting severe illness in COVID-19 patients remains challenging, lacking precise methods. This study aimed to leverage clinical real-world data and multiple machine-learning algorithms to formulate innovative predictive models for assessing the risk of severe outcomes or mortality in hospitalized patients with COVID-19. Data were obtained from the Taipei Medical University Clinical Research Database (TMUCRD) including electronic health records from three Taiwanese hospitals in Taiwan. This study included patients admitted to the hospitals who received an initial diagnosis of COVID-19 between January 1, 2021, and May 31, 2022. The primary outcome was defined as the composite of severe infection, including ventilator use, intubation, ICU admission, and mortality. Secondary outcomes consisted of individual indicators. The dataset encompassed demographic data, health status, COVID-19 specifics, comorbidities, medications, and laboratory results. Two modes (full mode and simplified mode) are used; the former includes all features, and the latter only includes the 30 most important features selected based on the algorithm used by the best model in full mode. Seven machine learning was employed algorithms the performance of the models was evaluated using metrics such as the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, and specificity. The study encompassed 22,192 eligible in-patients diagnosed with COVID-19. In the full mode, the model using the light gradient boosting machine algorithm achieved the highest AUROC value (0.939), with an accuracy of 85.5%, a sensitivity of 0.897, and a specificity of 0.853. Age, vaccination status, neutrophil count, sodium levels, and platelet count were significant features. In the simplified mode, the extreme gradient boosting algorithm yielded an AUROC of 0.935, an accuracy of 89.9%, a sensitivity of 0.843, and a specificity of 0.902. This study illustrates the feasibility of constructing precise predictive models for severe outcomes or mortality in COVID-19 patients by leveraging significant predictors and advanced machine learning. These findings can aid healthcare practitioners in proactively predicting and monitoring severe outcomes or mortality among hospitalized COVID-19 patients, improving treatment and resource allocation.

Posterior Müller Muscle-Conjunctival Resection as a First Step to Treat Eyelid Ptosis: Clinical Results and Treatment Algorithm.

The posterior Müller muscle-conjunctival resection (MMCR) procedure is a straightforward procedure for the correction of eyelid ptosis with a relatively short operating time and fast recovery. Traditionally, its use was limited to patients with mild involutional ptosis and good levator function and a positive phenylephrine test result. To evaluate the efficacy of the MMCR procedure as a primary step to treat eyelid ptosis with varying etiology and severity, including patients with more severe ptosis and moderate levator function, and to produce a treatment algorithm. A retrospective analysis of the results of 34 patients, comprising 56 operated eyelids, treated with the MMCR procedure for eyelid ptotis between 2016 and 2018, was performed. Preoperative and postoperative pictures were analyzed for determining the margin-to-reflex distance (MRD1), symmetry, and complications. We found a mean preoperative MRD1 of 1.3 mm (SD 1.1) and postoperative MRD1 of 3.2 mm (SD 1.0). The mean postoperative MRD1 for unilateral and bilateral cases was 3.4 (SD 0.8) and 3.2 (SD 1.1), respectively. Only 2 patients (5.9%) had an asymmetrical postoperative result (>1.0 mm MRD1 difference), and both were unilateral cases. Complications were scarce: only 1 patient (2.9%) developed dry eyes and 2 patients experienced temporary discomfort from the conjunctival sutures. The MMCR procedure appears to be an excellent procedure as a primary step to correct eyelid ptosis with varying etiologies and severity, due to its low risk of asymmetry, short learning curve, and high success rate. A flow chart as treatment algorithm is provided for clinical decision making.

Reassessing the Use of Race in Clinical Algorithms: An Interactive, Case-Based Session for Medical Students Using eGFR.

Medical curricula implicitly teach that race has a biological basis. Clinical rotations reinforce this misconception as race-based algorithms are used to guide clinical decision-making. This module aims to expose the fallacy of race in clinical algorithms, using the estimated glomerular filtration rate (eGFR) equation as an example. We created a 60-minute module in consultation with nephrologists. The format was an interactive, case-based presentation with a didactic section. A third-year medical student facilitated the workshops to medical students. Evaluation included pre/post surveys using 5-point Likert scales to assess awareness regarding use of race as a biological construct. Higher scores indicated increased awareness. Fifty-five students participated in the module. Pre/post results indicated that students significantly improved in self-perceived knowledge of the history of racism in medicine (2.6 vs. 3.2, p < .001), awareness of race in clinical algorithms (2.7 vs. 3.7, p < .001), impact of race-based eGFR on quality of life/treatment outcomes (4.5 vs. 4.8, p = .01), differences between race and ancestry (3.7 vs. 4.3, p < .001), and implications of not removing race from the eGFR equation (2.7 vs. 4.2, p < .001). Students rated the workshops highly for quality and clarity. Our module expands on others' work to expose the fallacy of race-based algorithms and define its impact on health equity. Limitations include a lack of objective assessment of knowledge acquisition. We recommend integrating this module into preclinical and clinical curricula to discuss the use of race in medical literature and clinical practice.

Optimization of the clinical diagnostic examination algorithm of patients with a cross bite complicated by cranio-mandibular dysfunction and postural disorder.

Aim: Optimization of the clinical and diagnostic examination algorithm of patients with cross bite, aggravated by cranio-mandibular dysfunction and postural disorders. Materials and Methods: 22 patients aged 13-16 years with cross bite with displacement of the lower jaw were examined. The first group consisted of 15 people with a right-sided displacement of the lower jaw, the second - 7 patients with a left-sided one. The condition of the musculoskeletal system was assessed by the position of the head, shoulders, shoulder blades, back (curvature of the spine), legs, chest shape, and abdomen. To determine the state of stability of the body in space, posturological and kinesiological tests were performed. The location of TMJ elements was evaluated on orthopantomograms. Statistical processing of the material was carried out with the help of the "Excel" license package. Results: 63.64% of patients with a cross bite have disorders of the musculoskeletal system: scoliotic posture - 40.91% and scoliosis - 22.73%. TMJ dysfunction was detected in all examined patients. It was established that the anatomical and topographic features of the joint elements depend on the side of the lower jaw displacement. Conclusions: The functional imbalance of all structural elements of the musculoskeletal system and the cranio-mandibular complex determined during the research proved the need to optimize the clinical-diagnostic algorithm: consultation of a traumatologist-orthopedic doctor, X-ray examination of TMJ, conducting posturological tests.

Enhancing Staff Safety in Swallowing Assessment: Insights Post COVID Pandemic

The COVID-19 pandemic was a time of significant challenge in the healthcare industry. In particular, hospitals worked diligently to develop protocols and guidelines to reduce viral transmission to both patients and staff. Service delivery concerning the instrumental assessment of swallowing evaluation was particularly challenging during the pandemic given the high risk of viral transmission to clinical staff. Clinical algorithms were developed with the goal of implementing a measured return to service delivery while preserving the clinical workforce. The conceptual development and evolution of the algorithms are discussed to highlight knowledge gained from the COVID-19 pandemic with the objective of assisting other professionals who perform essential functions in ensuring their safety and mitigating virus transmission. Should another healthcare crisis arise, insights from this experience can inform other institutions who want to prioritize both preservation of workforce and delivery of evidence-based care.

Management of patients with insomnia and polymorbidity. A draft of the clinical guidelines

Insomnia is a serious and widespread public health problem, but is often undetected and patients do not receive needed treatment. Insomnia is often comorbid with other diseases and conditions, such as arterial hypertension, type 2 diabetes mellitus, pain syndromes, anxiety and depressive disorders, etc. A separate problem is drug-induced insomnia, when patients develop symptoms due to other diseases treatments. Insomnia has a negative effect on the prognosis of comorbid diseases, including an increased risk of death, more severe disease, and decreased quality of life. The presence of sleep disorders makes it difficult to effectively treat the underlying disease, so clinical guidelines draft for the evaluation and treatment of insomnia in multimorbid patients is proposed. Diagnostic methods are reviewed and recommendations are given for the treatment of acute and chronic insomnia and features of the treatment of insomnia in multimorbid patients. A clinical algorithm has been proposed to determine treatment tactics in multimorbid patients.

Clinical Algorithm and Diagnostic Tools for Lyme Disease.

Lyme disease (LD) is the prototype of tick-borne infections. The broad spectrum of LD symptomatology, together with the tremendous variety of sensibility and specificity of diagnostic tests, poses a complex challenge for LD diagnosis. Here, we propose a clinical algorithm for Lyme patients to prevent treatment delay in suspicious scenarios.

Young Infants Clinical Signs Study 8-sign Algorithm for Identification of Sick Infants Adapted for Routine Home Visits: A Systematic Review and Critical Appraisal of its Measurement Properties.

Objective. The 8-sign algorithm adapted from the Young Infants Clinical Signs Study (YICSS) is widely used to identify sick infants during home visits (YICSS-home algorithm). We aimed to critically appraise the development and evidence of measurement properties, including sensibility, reliability, and validity, of the YICSS-home algorithm. Methods. Relevant studies were identified through a systematic literature search. Results. The YICSS-home algorithm has good sensibility. The algorithm demonstrated at least moderate inter-rater reliability and sensitivity ranging from 69% to 80%. However, the algorithm was developed among sick infants brought for care to a health facility and not initially developed for use by community health workers (CHWs) during home visits. Some important risk factors were omitted at item generation. Inter-CHW reliability and construct validity have not been estimated. Conclusion. Future research should build on the strengths of the YICSS-home algorithm and address its limitations to develop a new algorithm with improved predictive accuracy.

Socio-ethical challenges and opportunities for advancing diversity, equity, and inclusion in digital medicine.

Digitalization in medicine offers a significant opportunity to transform healthcare systems by providing novel digital tools and services to guide personalized prevention, prediction, diagnosis, treatment and disease management. This transformation raises a number of novel socio-ethical considerations for individuals and society as a whole, which need to be appropriately addressed to ensure that digital medical devices (DMDs) are widely adopted and benefit all patients as well as healthcare service providers. In this narrative review, based on a broad literature search in PubMed, Web of Science, Google Scholar, we outline five core socio-ethical considerations in digital medicine that intersect with the notions of equity and digital inclusion: (i) access, use and engagement with DMDs, (ii) inclusiveness in DMD clinical trials, (iii) algorithm fairness, (iv) surveillance and datafication, and (v) data privacy and trust. By integrating literature from multidisciplinary fields, including social, medical, and computer sciences, we shed light on challenges and opportunities related to the development and adoption of DMDs. We begin with an overview of the different types of DMDs, followed by in-depth discussions of five socio-ethical implications associated with their deployment. Concluding our review, we provide evidence-based multilevel recommendations aimed at fostering a more inclusive digital landscape to ensure that the development and integration of DMDs in healthcare mitigate rather than cause, maintain or exacerbate health inequities.

A personalized approach to recurrent abdominal pain syndrome based on clinical and laboratory algorithms

To optimize the diagnosis of abdominal pain syndrome occurring under the «mask» of acute pancreatitis via algorithms for clinical and laboratory examination. We have retrospectively and prospectively analyzed patients with abdominal pain syndrome. We selected all patients with acute pancreatitis who repeatedly applied with abdominal pain syndrome to polyclinics and hospitals between 2017 and 2021. A personalized algorithm for patients with abdominal pain syndrome occurring under the «mask» of acute pancreatitis has been developed. This algorithm underlies an information system for decision-making support. An optimal diagnostic algorithm is needed in reception departments of hospitals and polyclinics. This one depends on equipment of hospitals and needs to be constantly improved. When refusing hospitalization and discharging patients with recurrent pain syndromes, physicians should orient the doctors of general network to use more informative methods. Indeed, the last ones may be unavailable in reception departments and various districts due to technical, qualification and organizational shortcomings. Advanced clinical and laboratory diagnostic methods based on the above-described algorithm are necessary for recurrent abdominal pain syndrome occurring under the «mask» of acute pancreatitis.

Body Mass Index Is Associated With Myocutaneous Free Flap Reliability: Overcoming the Obesity Obstacle With a Proposed Clinical Algorithm to Identify and Manage High-Risk Patients Undergoing Gracilis Free Flap With Skin Paddle Harvest.

The purpose of this study was to evaluate the role of body mass index (BMI) in predicting postoperative complications following myocutaneous free flap transfer. In addition, we sought to identify certain body composition variables that may be used to stratify patients into low- versus high-risk for gracilis myocutaneous free flap with skin paddle failure. Using the National Surgical Quality Improvement Program database, we collected data for all patients who underwent myocutaneous free flap transfer from 2015 to 2021. Demographic data, medical history, surgical characteristics, and postoperative outcomes, including complications, reoperations, and readmissions, were collected. Body mass index was correlated with outcome measures to determine its role in predicting myocutaneous free flap reliability. Subsequently, we retrospectively obtained measurements of perigracilis anatomy in patients who underwent computed tomography angiography bilateral lower extremity scans with intravenous contrast at our institution. We compared body composition data with mathematical equations calculating the potential area along the skin of the thigh within which the gracilis perforator may be found. Across the United States, 1549 patients underwent myocutaneous free flap transfer over the 7-year study period. Being in obesity class III (BMI ≥40 kg/m2) was associated with a 4-times greater risk of flap complications necessitating a return to the operating room compared with being within the normal BMI range. In our computed tomography angiography analysis, average perigracilis adipose thickness was 18.3 ± 8.0 mm. Adipose thickness had a strong, positive exponential relationship with the area of skin within which the perforator may be found. In our study, higher BMI was associated with decreased myocutaneous free flap reliability. Specifically, inner thigh adipose thickness can be used to estimate the area along the skin within which the gracilis perforator may be found. This variable, along with BMI, can be used to identify patients who are considered high-risk for flap failure and who may benefit from additional postoperative monitoring, such as the use of a color flow Doppler probe and more frequent and prolonged skin paddle monitoring.

Scrub Typhus Diagnostics: The Present and the Future

Scrub typhus, caused by Orientia tsutsugamushi, is a vector-borne, zoonotic disease commonly occurring in a geographic region known as the ‘tsutsugamushi triangle’. Scrub typhus causes acute undifferentiated febrile illness (AUFI) with non-specific clinical features and is difficult to diagnose when a highly characteristic but not pathognomonic eschar is absent. Because the eschar is not always present, laboratory tests are required for diagnosis. Serological assays have been the mainstay of laboratory diagnosis of scrub typhus to date. Here, we present the major clinical features and clinical algorithms suggestive of this disease, to aid in better selection of diagnostic methods. The advantages and disadvantages of various scrub typhus assays are also discussed. Furthermore, we describe diagnostics, including serological and molecular assays, that may be available in the near future.

DISCUSSION ON INDIVIDUAL EDUCATIONAL TRAJECTORY AS AN INTEGRAL COMPONENT OF CONTINUOUS PROFESSIONAL DEVELOPMENT

The system of continuous professional development of doctors in Ukraine needs detailed analysis and comprehensive dynamic monitoring to become a process that includes the possibility of a doctor’s professional growth. The research aims to monitor approaches to the continuous professional development of dentists for the formation and improvement of professional competence skills by highlighting the role of individual educational trajectory based on the priority of work formats with professional information, methods of familiarization with it and its receiving. 2216 respondents took part in the research, filling out a questionnaire during scientific and practical events of professional schools for dentists in online mode in March 2023, which included questions on the definition and objectification of forms, formats, ways of implementing continuous professional development and the vision of approaches to acquiring new knowledge and skills. The data obtained was processed using the StatSoft Inc licence program. Analytical research methods and structural-logical analysis were applied. According to the survey results, an individual's educational trajectory should include attending lectures, professional seminars, working with scientific and practical publications. The main quality requirement for publications is their good illustration of clinical cases and algorithms for working with patients. The analysis results of the rating parameters of professional dental journals declare the request of the dental community to improve the quality of material content, visual design and a possible change in positioning.

PREDICTING DEMENTIA DIAGNOSIS FROM COGNITIVE FOOTPRINTS IN ELECTRONIC MEDICAL RECORDS USING MACHINE LEARNING

Abstract The cognitive footprint theory suggests that a person’s risk of dementia is affected by activities and events (footprints) across the lifespan. This study aimed to develop a clinical algorithm for predicting dementia diagnosis from cognitive footprints in electronic medical records (EMRs) using machine learning. Population-based EMRs from the Clinical Data Analysis and Reporting System in Hong Kong were employed. We included all patients with dementia diagnosed at 65+ between 2000 and 2018. Controls without dementia were matched (1:1) by age, sex, and index date. The prediction accuracy of a standard logistic regression model was compared with five machine learning models (LASSO, Random Forest, Multilayer perceptron, XGBoost, and LightGBM), using a wide range of established and exploratory risk factors identified at mid- ( &amp;lt; 65 years) and late-life (65+ years). A total of 261,116 individuals (59.3% female; mean age at index date [SD]: 82.70 [8.54]) with and without dementia were included. Compared with the logistic model (area under the curve [AUC]: 0.60), the predictive and diagnostic accuracy of dementia improved substantially in machine learning models. The Multilayer perceptron and LightGBM models showed comparable performance with the same test AUC of 0.720. LASSO (AUC: 0.765) and Random Forest (AUC: 0.768) showed a slightly worse performance. Antipsychotic drugs, education, antidepressants and head injury in late life were the top 4 important predictors in all models. The machine learning-based algorithm for predicting dementia can be used to identify patients with an increased likelihood of dementia to allow precise and timely primary interventions.