Roshan Koju

Work place: Nepal College of Information Technology



Research Interests: Image Processing, Image Manipulation, Image Compression, Artificial Intelligence


Dr. Roshan Koju: Lecturer, Nepal College of Information Technology. He completed his Ph.D. in Electronics and Computer Engineering from Tribhuwan University in 2018. His research areas include Artificial intelligence, data science, Image processing, Networking, and Wireless Communication.

Author Articles
Healthcare Vulnerability Mapping Using K-means ++ Algorithm and Entropy Method: A Case Study of Ratnanagar Municipality

By Apurwa Singh Roshan Koju

DOI:, Pub. Date: 8 Apr. 2023

Healthcare is a fundamental human right. Vulnerable populations in healthcare refer to those who are at greater risk of suffering from health hazards due to various socio-economic factors, geographical barriers, and medical conditions. Mapping of this vulnerable population is a vital part of healthcare planning for any region. Very few such research regarding the distribution of healthcare service providers was carried out in the Nepali context previously. Thus, the results of vulnerability mapping can help with meaningful interventions for healthcare demands. This study focused on combining geo-analytics, unsupervised machine learning algorithms, and entropy methods for performing vulnerability mapping. K-means++ clustering algorithm was applied to household data of Ratnanagar municipality for the purpose of creating multiple clusters of households. An open-source routing machine was used to compute the distance to the nearest health service provider from each household in Ratnanagar municipality. The entropy method was used to evaluate the vulnerability measure of each cluster. Later, based on the population of different clusters in each ward and their respective vulnerability measures, each ward’s vulnerability measure was quantified. It can be observed that wards that are farther away from the east-west highway have higher vulnerability indices. This study found that machine learning algorithms can be effectively used in combination with the weighting method for vulnerability mapping. Using an unsupervised machine learning algorithm made sure that dimensions of vulnerability are visible.

[...] Read more.
Other Articles