Background Thrombocytopenia is a common hematological disorder encountered in clinical practice. To improve diagnostic efficiency and patient outcomes, it is important to understand its frequent clinical manifestations and predominant etiologies in different healthcare settings. This study aimed to describe the typical clinical presentations and underlying causes of thrombocytopenia among patients at a tertiary care hospital in the Kingdom of Saudi Arabia. Study design A single-center, retrospective cross-sectional study. Materials and methods Patients with thrombocytopenia (platelet count < 150,000/microL) admitted between January and December 2022 were included. Patients with pseudothrombocytopenia were excluded. Demographic, clinical, and laboratory data were collected through manual review of electronic medical records. Etiologies were recorded, and statistical analysis was performed using Stata software. Here and throughout, n denotes the number of patients contributing data to a given comparison, and N denotes the total cohort (N = 114). Where applicable, denominators were adjusted for missing data. Results Of 114 patients, 73 (64.04%) were male. The mean age was 40.80 ± 15.93 years, and the mean length of hospital stay was 11.83 ± 15.73 days. The mean platelet count was 51,480 ± 39,690/microL. Mild thrombocytopenia was observed in 19 (16.67%) patients, moderate in 31 (27.19%), and severe in 64 (54.14%). Infection was the most common etiology (n = 48, 42.11%), with Plasmodium species accounting for the largest number of cases (n = 25). Bleeding manifestations were observed in 72 (63.07%) patients, most commonly epistaxis (n = 16, 14.04%), ecchymosis (n = 13, 11.4%), and petechiae (n = 9, 7.89%). Patients with bleeding, particularly petechiae, ecchymosis, epistaxis, and gum bleeding, had significantly lower platelet counts than those without bleeding symptoms. Conclusion Infections were the predominant cause of thrombocytopenia in this cohort. Lower platelet counts were significantly associated with bleeding manifestations, particularly epistaxis, emphasizing the need for early recognition and appropriate clinical management.
Wearable Sensor (WS)-based monitoring systems detect minute patient movements/ demands and abnormalities through periodic sensing and imaging. Sensor data observed over different intervals is not constant or available based on operating sequences. Due to variations in data sequences, the analysis process becomes complex, resulting in less precise outputs. To address this problem, an Allied Data Disparity Technique (ADDT) is proposed in this article. This technique identifies the disparity in different monitoring sequences in coherence with the clinical and previous values. Based on the mean disparity, the data requirement for the WS sequence is decided. This decision uses multiple substituted and predicted values obtained from previous instances. Multi-Instance Ensemble Perceptron Learning is used in this decision process, where the substitution instances for clinical and previous outcomes are performed. The ensemble perceptron selects the maximum clinical value correlating sensor data to ensure high sequence prediction. The ensembles are updated based on the highest precision-based WS values for diagnosis. This diagnosis-focused coalition between clinical and predicted WS values is updated periodically for the new precision levels identified.
wearable sensorspatient health monitoringclinical data fusionmissing data imputation
In recent times, the global rise in the prevalence rate of amyotrophic lateral sclerosis (ALS) has profoundly affected the welfare of several people in the world. ALS is a lethal neurodegenerative disease (NDD) that damages the nerve cells in the brain and spinal cord. Moreover, it removes the person’s capability of controlling muscle movements in the body. It is necessary to detect the disease earlier, to reduce the disease severity, and to enhance the life expectancy of the patients. Traditionally, ALS screening is handled by qualified physicians through blood tests, which is an expensive, painful, and time-consuming process. To resolve this limitation, several researchers focused on the ALS classification. Conversely, it have a few drawbacks, such as lack of accuracy and speed, overfitting of data, and noise handling tasks. For enhancing the classification of ALS, the proposed approach employs progressive entropy weighted-based focal loss (PEWFL)-XGBoost through the Kaggle ALS dataset. The XGBoost is used for the ability to manage missing data and speed. Nevertheless, it has certain limitations such as overfitting of data, hyperparameter tuning, and handling of smaller datasets. To resolve this, PEWFL is added to the XGBoost system to improve the classification performance. Correspondingly, the efficiency of the respective system is calculated using performance metrics to evaluate the performance of the research. Moreover, internal comparison with classical algorithms such as XGBoost, K-nearest neighbor, and random forest discloses the efficacy of the respective model. The proposed system is envisioned to contribute to molecular genetics and neuroscience research and assist neurologists in enhancing the diagnosis of ALS.
In the face of an aging population, smart healthcare services are now within reach, thanks to the proliferation of high-speed internet and other forms of digital technology. Data problems in smart healthcare, unfortunately, put artificial intelligence in this area to serious limitations. There are several issues, including a lack of standard samples, noisy data interference, and actual data that is missing. A three-stage AI-based data generating strategy is suggested to handle missing datasets, using a small sample dataset obtained from a smart healthcare program community in a specific city: Step one involves generating the dataset's basic attributes using a tree-based generation strategy that takes the original data distribution into account. Step two involves using the Naive Bayes algorithm to create basic indicators of behavioural capability assessment for the samples. Step three builds on stage two and uses a multivariate linear regression method to create evaluation criteria and indicators of high-level behavioural capability. Six problems involving multiple classifications and two tasks using multiple labels are implemented using various neural network-based training strategies on the obtained data to assess the usefulness of the dataset for downstream tasks. To ensure that the data collected is genuine and useful, the experimental data must be analysed and expert knowledge must be included.
synthetic data generationmissing data imputationsmart healthcaredisease prevention
BULLETIN OF STOMATOLOGY AND MAXILLOFACIAL SURGERY · 132-147
Background: Artificial Intelligence (AI) in dentistry has the potential to revolutionize oral healthcare by solving its inherent shortcomings. Aim: To review and evaluate the body of research on artificial intelligence's use in dentistry, with a focus on how it affects treatment planning, diagnosis, and patient care in a range of dental specialties. Methodology: 30 papers encompassing oral diagnosis, surgery, endodontics, prosthodontics, orthodontics, forensic dentistry, radiography, and periodontics are thoroughly examined in this review using PRISMA guidelines. The Cochrane Handbook principles were followed in the evaluation of important variables such as randomization, blinding, withdrawal/dropout rates, sample size estimation, clarity of inclusion/exclusion criteria, examiner reliability testing, pre-specification of outcomes, and bias risk. The Newcastle-Ottawa Scale (NOS) was used in quality assessment to measure bias risk and star ratings. Results: The research highlight improvements in diagnosis, treatment planning, and procedural accuracy, illustrating the revolutionary effects of AI in dentistry. Applications of AI demonstrate its versatility and include automated designs, risk prediction, lesion recognition, and precision in dental operations. There is little chance of bias in randomization, intervention variations, and outcome assessments, according to the methodological evaluation, which shows excellent scientific rigor. Even though a few studies had minor issues including uneven blinding and missing data, these had no appreciable impact on the dependability of the results. Overall, the studies' consistent methodological quality highlights how AI may be relied upon to advance dental research and practice.
artificial intelligence in dentistrydental diagnosisdental treatment planningendodontics
Stroke survivors have access to a wide range of drug and non-drug treatments for the resulting physiological and functional problems. However, comprehensive therapies typically fail to meet the demands of a large percentage of patients. The recent clinical studies to improve protocol scientific evidence have resulted in a new development phase for rehabilitation medicine. Stroke rehabilitation supports individuals to lead a normal life. It assists the physicians in offering an effective environment to the patients. The evaluation of a patient’s progress in rehabilitation is based on the clinician’s subjective observations and the patient’s self-reported data. Deep learning techniques offer novel forms of individualized treatment. Nonetheless, missing data is one of the crucial factors that reduces the performance of data classification techniques. Thus, there is a demand for functional recovery prediction models for supporting stroke patients (SPs) to improve their quality of life. In this study, the researchers intend to build a framework for predicting functional outcomes using the electronic health record data of SPs. An attention-based bidirectional gated recurrent unit is used for developing the data imputation model. In addition, a shallow-convolutional neural network is employed for predicting the functional outcomes based on the modified Barthel Index. Data from 356 SPs were utilized for evaluating the performance of the proposed framework with the benchmark metrics and baseline models. The findings reveal that the proposed framework outperforms the state-of-the-art classification by achieving an average accuracy, precision, recall, F1-measure, specificity, and sensitivity of 98.18, 97.48, 98, 97.74, 96.74, and 97.24, respectively. The proposed framework can be implemented in real time to support SPs.
stroke rehabilitationfunctional recovery predictionelectronic health recordsmissing data imputation