Analysis and Prediction of Diabetes Complication Disease using Data Mining Algorithm

The power of light and sound: optoacoustic skin imaging for diabetes progression monitoring

Diabetes progression is marked by damage to vascular and neural networks. Raster-scan optoacoustic mesoscopy holds the potential to measure extent of diabetes progression by analyzing changes in skin vasculature.

Early detection system of risk factors for diabetes mellitus type 2 utilization of machine learning-random forest

BACKGROUND

The prevalence of morbidity and mortality for type 2 diabetes mellitus (DM) is still increasing because of changing lifestyles. There needs to be a means of controlling the rise in the incidence of the disease. Many researchers have utilized technological advances such as machine learning for disease prevention and control, especially in noncommunicable conditions. Researchers are, therefore, interested in creating an early detection system for risk factors of type 2 diabetes.

MATERIALS AND METHODS

The study was conducted in February 2022, utilizing secondary surveillance data from Puskesmas Johar Baru, Jakarta, in 2019, 2020, and 2021. Data was analyzed utilizing various bivariate and multivariate statistical methods at 5% significance level and machine learning methods (random forest algorithm) with an accuracy rate of >80%. The data for the three years was cleaned, normalized, and merged.

RESULTS

The final population was 65,533 visits out of the initial data of 196,949, and the final number of DM 2 population was 2766 out of the initial data of 9903. Age, gender, family history of DM, family history of hypertension, hypertension, high blood sugar levels, obesity, and central obesity were significantly associated with type 2 DM. Family history was the strongest risk factor of all independent variables, odds ratio of 15.101. The classification results of feature importance, with an accuracy rate of 84%, obtained in order were age, blood sugar level, and body mass index.

CONCLUSION

Blood sugar level is the most influential factor in the incidence of DM in Puskesmas Johar Baru. In other words, a person with a family history of type 2 diabetes, at unproductive age, of female gender, and of excessive weight can avoid type 2 diabetes if they can regularly maintain their blood sugar levels.

Machine Learning Modeling of Disease Treatment Default: A Comparative Analysis of Classification Models

Generally, treatment default of diseases by patients is regarded as the biggest threat to favourable disease treatment outcomes. It is seen as the reason for the resurgence of infectious diseases including tuberculosis in some developing countries. Sadly, its occurrence in chronic disease management is associated with high morbidity and mortality rates. Many reasons have been adduced for this phenomenon. Exploration of treatment default using biographic and behavioral metrics collected from patients and healthcare providers remains a challenge. The focus on contextual nonbiomedical measurements using a supervised machine learning modeling technique is aimed at creating an understanding of the reasons why treatment default occurs, including identifying important contextual parameters that contribute to treatment default. The predicted accuracy scores of four supervised machine learning algorithms, namely, gradient boosting, logistic regression, random forest, and support vector machine were 0.87, 0.90, 0.81, and 0.77, respectively. Additionally, performance indicators such as the positive predicted value score for the four models ranged between 98.72%–98.87%, and the negative predicted values of gradient boosting, logistic regression, random forest, and support vector machine were 50%, 75%, 22.22%, and 50%, respectively. Logistic regression appears to have the highest negative-predicted value score of 75%, with the smallest error margin of 25% and the highest accuracy score of 0.90, and the random forest had the lowest negative predicted value score of 22.22%, registering the highest error margin of 77.78%. By performing a chi-square correlation statistic test of variable independence, this study suggests that age, presence of comorbidities, concern for long queuing/waiting time at treatment facilities, availability of qualified clinicians, and the patient’s nutritional state whether on a controlled diet or not are likely to affect their adherence to disease treatment and could result in an increased risk of default.

An effective integrated machine learning approach for detecting diabetic retinopathy

Millions of people across the world are suffering from diabetic retinopathy. This disease majorly affects the retina of the eye, and if not identified priorly causes permanent blindness. Hence, detecting diabetic retinopathy at an early stage is very important to safeguard people from blindness. Several machine learning (ML) algorithms are implemented on the dataset of diabetic retinopathy available in the UCI ML repository to detect the symptoms of diabetic retinopathy. But, most of those algorithms are implemented individually. Hence, this article proposes an effective integrated ML approach that uses the support vector machine (SVM), principal component analysis (PCA), and moth-flame optimization techniques. Initially, the ML algorithms decision tree (DT), SVM, random forest (RF), and Naïve Bayes (NB) are applied to the diabetic retinopathy dataset. Among these, the SVM algorithm is outperformed with an average of 76.96% performance. Later, all the aforementioned ML algorithms are implemented by integrating the PCA technique to reduce the dimensions of the dataset. After integrating PCA, it is noticed that the performance of the algorithms NB, RF, and SVM is reduced dramatically; on the contrary, the performance of DT is increased. To improve the performance of ML algorithms, the moth-flame optimization technique is integrated with SVM and PCA. This proposed approach is outperformed with an average of 85.61% performance among all the other considered ML algorithms, and the classification of class labels is achieved correctly.

Effect of Artemisia extract on glycemic control: A systematic review and meta-analysis of randomized controlled trial

BACKGROUND AND AIM

Animal and human studies have indicated anti-diabetic effect of Asteraceae. The present study aimed to systematically review and analyse randomized controlled trials assessing the effect of Artemisia extract on glycemic status in patients with impaired glycemic control.

METHODS

Web of Science, Cochrane library, EMBASE and PubMed databases were searched from the earliest possible date up to 7th October 2020.

RESULTS

Six studies were included in the meta-analysis. Analysis showed that supplementation with Artemisia extract decreased homeostatic model assessment of insulin resistance (HOMA-IR) (-0.734, 95% CI: -1.236 to -0.232, P = .019) in comparison to placebo. However, reductions in fasting blood glucose (FBG) (-0.595, 95% CI: -1.566 to 0.376, P = .164), insulin (-0.322, 95% CI: -1.047 to 0.404, P = .286) and glycated haemoglobin (-0.106, 95% CI: -0.840 to 0.629, P = .678) were not statistically significant.

CONCLUSION

Supplementation with Artemisia extract may reduce HOMA-IR, but beneficial effects on other markers such as FBG requires further investigations.

IoT Based Health—Related Topic Recognition from Emerging Online Health Community (Med Help) Using Machine Learning Technique

The unprompted patient’s and inimitable physician’s experience shared on online health communities (OHCs) contain a wealth of unexploited knowledge. Med Help and eHealth are some of the online health communities offering new insights and solutions to all health issues. Diabetes mellitus (DM), thyroid disorders and tuberculosis (TB) are chronic diseases increasing rapidly every year. As part of the project described in this article comments related to the diseases from Med Help were collected. The comments contain the patient and doctor discussions in an unstructured format. The sematic vision of the internet of things (IoT) plays a vital role in organizing the collected data. We pre-processed the data using standard natural language processing techniques and extracted the essential features of the words using the chi-squared test. After preprocessing the documents, we clustered them using the K-means++ algorithm, which is a popular centroid-based unsupervised iterative machine learning algorithm. A generative probabilistic model (LDA) was used to identify the essential topic in each cluster. This type of framework will empower the patients and doctors to identify the similarity and dissimilarity about the various diseases and important keywords among the diseases in the form of symptoms, medical tests and habits.