An Optimised Multivariable Regression Model for Predictive Analysis of Diabetic Disease Progression

Abstract

With the advent of smart systems and smart IoT network all over the world leading to enormous amount of data generation; the right analysis and decision making based on the relevant data plays a crucial role. Various industries such as transportation, retail, healthcare etc. rely on analysis using this huge volumes of data for intelligent decision making. In smart healthcare system, accurate analysis of patients' data and prediction of diseases and medicine is important. To a great extent, fatalities can be avoided by timely recommendation of healthcare measures and immediate alert on emergency conditions. The use of machine learning algorithm for precise predictive analysis of data can be very promising in the field of healthcare. In this paper, optimised Multivariable Linear regression method is used to predict the diabetic disease progression of 442 patients based on various parameters such as age, gender, Body Mass Index and 6 different blood serum measurements. Here optimisation is performed using feature reduction and logarithmic transformation. The predicted output is found to be closely associated with actual output data with a Root Mean Square Error of 1.5 units; which indicates higher accuracy in comparison with the non-optimised model with the error of 54 units. There has also been a comparison with the results obtained from other state of the art regression methods, which proves that the proposed model exhibits maximum accuracy. This method can be used to provide promising medical advice to the patients on how to reduce the diabetic disease progression over a year by controlling various health parameters.

Data-driven sales optimization with regression and chaotic pattern search

“…Step 4: To the pivoted table of Step 3, apply the ordinary least square (OLS) model of regression ( Daliya, Ramesh & Ko, 2021 ; Sathyadevan & Chaitra, 2015 ) by taking the number of days (second column of Table 8 ) as dependent and all paths (remaining columns of Table 8 ) as independent variables. The result of the OLS model gives an approximation for the coefficients of ’s, i.e ., ’s in the objective function along with a standard error of regression, .…”

High Sensitivity Envelope Detector Assisted by Microwave Photonics

“…Each of these forms of error analysis has a higher or lower sensitivity regarding the variation of values, as a result, when possible, it is interesting to use more than one form of evaluation. As shown in Equations 20 to 24, respectively, (Daliya et al, 2021). where, 𝑌 𝑖 is the measured value, 𝑌 𝑖 ̂ is the value predicted by the models analyzed and n is the sample size.…”

“…where, 𝑌 𝑖 is the measured value, 𝑌 𝑖 ̂ is the value predicted by the models analyzed and n is the sample size. Table 1 shows the results for error analysis using the techniques of MAE, MSE, RMSE, and MSLE (Daliya et al, 2021).…”

Healthcare monitoring and Recommendation model: Hybrid Deep learning Enabled Disease Diagnosis Framework

“…Consequently, investigators and physicians need to acknowledge the importance of ML approaches. Several approaches were devised and documented over the past years for predicting the diabetic-cardio [14] [15] [16]. Though risk forecasting strategies are available, the majority among them consider just a selection of risk variables.…”