Search

This study aims to identify telecom customer segments by utilizing machine learning and subsequently develop a web-based dashboard. The dashboard visualizes the cluster analysis based on demographics, behavior, and region features. The study applied analytic pipeline that involved five stages i.e. data generation, data pre-processing, data clustering, clusters analysis, and data visualization. Firstly, the customer’s dataset was generated using Faker Python package. Secondly was the pre-processing which includes the dimensionality reduction of the dataset using the PCA technique and finding the optimal number of clusters using the Elbow method. Unsupervised machine learning algorithm K-means was used to cluster the data, and these results were analyzed and labeled with labels and descriptions. Lastly, a dashboard was developed using Microsoft Power BI to visualize the clustering results in meaningful analysis. According to the results, four customer clusters were obtained. An interactive web-based dashboard called INSIGHT was developed to provide analysis of customer segments based on demographic, behavioral, and regional traits; and to devise customized query for deeper analysis. The correctness of the clustering results was evaluated and achieved a satisfactory Silhouette Score of 0.3853. Hence, the telecom could target their customers accurately based on their needs and preferences to increase service satisfaction.

Background: Unplanned readmission incurs additional cost to patients and contributes to the rising healthcare cost of our nation. Although numerous studies have investigated the predicting factors that contribute to hospital readmission, the majority of the studies focused on clinical and patient-related factors, and were not from the perspectives of clinicians such as patient navigators (PNs). Aim: To understand factors that predict patients’ readmission risks from the PNs’ perspectives. Method: Using purposive sampling, PNs with a minimum of 10 years of clinical experience in the adult acute-care setting participated in the focus group interviews. Thematic analysis was adopted. Findings: All 10 PNs agreed that the readmission risk assessment tool was useful as a guide to assess patients’ risk of readmission; however, they also mentioned the use of clinical judgement and experience while assessing their patients. Three themes emerged from this study: (1) looking beyond medical-related issues; (2) social and community support; (3) functional status of patients. Conclusion: Predicting patients’ risk of readmission is closely tied to the patients’ current medical conditions and caregiving needs. Ensuring individualised readmission risk assessment and identifying social issues early are key in ensuring a holistic discharge planning.

Objective: The value of ventilation-perfusion (VQ) single photon emission tomography/computed tomography (SPECT/CT) lobar quantification for pre-operative assessment of lobectomy and lung volume reduction is known. Our in-house developed software, RAH ventilation perfusion SPECT/CT quantification (RAHVQSQ) has been shown to be able to identify the target lobe for collapse in bronchoscopic lung volume reduction (BLVR) for advanced emphysema. We have proven inter and intra observer reproducibility but are yet to validate the accuracy of our program. This study aims to validate the accuracy of our quantitative program through comparison with a modified version of GE Q lung which is a commercial program certified for clinical use., Subjects and Methods: Ventilation-perfusion SPECT/CT data of 19 subjects from our previous study using RAHVQSQ for BLVR assessment were re-analysed using Q lung by 2 technologists independently and in a blinded fashion to determine lobar differential ventilation, perfusion and volume percentages. The data were from GE Hawkeye 4 and external CT, thus a modified version of Q lung was used. To determine interobserver variation in the 3 parameters between the 3 assessors, intraclass correlation coefficient (ICC) and Bland-Altman limits of agreement (LoA) were generated., Results: Paired comparisons between the 3 assessors had high ICC (range for ventilation: 0.69-0.97; perfusion: 0.69-0.97; volume: 0.63-0.97) and means of LoA differences close to zero (range for ventilation: -0.04 - 0.10; perfusion: 0.00-0.02; volume: -0.12 - 0.09) were noted indicative of good concordance for all parameters., Conclusion: Using VQ SPECT/CT data of participants with advanced airway disease, our study has found a close concordance of estimated differential lobar ventilation, perfusion and volume percentages using RAHVQSQ when compared with a duplicated blinded assessment using Q lung. The good concordance supports the validity of our quantitative methodology.

Background: To develop a simple method for simultaneous solid and liquid gastric emptying assessment using a dual isotope labelled breath test., Material and Methods: 13 patients were given 100 g ground beef labelled with 25 MBq (99m)Tc sulphur colloid and 74 KBq (14)C octanoic acid, and 150 ml 10% glucose drink labelled with 8 MBq (67)Ga citrate and 150 mg (13)C acetate. 10 normal volunteers were given the same test meals but labelled with (14)C and (13)C only. Breath was collected at baseline and regularly for 4 hours. The (14)CO(2) and (13)CO(2) activity was measured with liquid scintillation counting and mass spectroscopy. The times to maximum (14)CO(2) and (13)CO(2), were determined. Comparison was made between times to maximum (14)CO(2) with scintigraphic retention of (99m)Tc at 100 minutes and times to maximum (13)CO(2) with the scintigraphic half-clearance time of (67)Ga., Results: For the solid meal, the times to maximum (14)CO(2) were: 60-120 minutes in the 8 patients with normal gastric emptying of (99m)Tc; 75-145 minutes for the 10 healthy volunteers; 75-180 minutes for the remaining 5 patients with abnormal gastric emptying of (99m)Tc. There was a weak but significant correlation (r = 0.56, p < 0.025) between the time to maximum (14)CO(2) and gastric retention of (99m)Tc at 100 minutes. For the liquid meal, times to maximum (13)CO(2) were: 20-35 minutes for the 4 with normal gastric emptying of (67)Ga; 15-40 minutes for the 10 healthy volunteers; 20-75 minutes for the remaining 9 patients with abnormal gastric emptying of (67)Ga. There was a strong and significant correlation (r = 0.88, p < 0.005) between times to maximum (13)CO(2) and gastric half-clearance time of (67)Ga., Conclusions: Breath tests utilising test meals labelled with *C isotopes are valid alternatives to scintigraphic studies using (99m)Tc and (67)Ga for the simultaneous assessment of gastric emptying of solids and liquids.