ISSN 2348-1196 (print) International Journal of Computer Science and Information Technology Research ISSN 2348-120X (online) Vol. 8, Issue 2, pp: (5-9), Month: April - June 2020, Available at: www.researchpublish.com
Patient-Case Similarity 1
Vishakha Sharma, 2Tannya Mittal, 3Shrey Shukla, 4Tushar Babbar, 5 Ms. Priyanka Bhardwaj Department Of Computer Science and Engineering MIET, Meerut
{vishakha.sharma.cs.2016, tannya.mittal.cs.2016, shrey.shukla.cs.2016, tushar.babbar.cs.2016, priyanka.bhardwaj}@miet.ac.in
Abstract: The aim of patient-case similarity is to recognize identical patients on the basis of their medical reports. Recognition of such cases of patients can be used in improving the results of drug proposal for new patients, clinical decision support, prediction of clinical outcome, research on these cases. We have applied machine learning algorithm to accomplish the specified aim with higher accuracy. Keywords: patient-case, word2vec, vectorizing.
1. INTRODUCTION In this era of high number of patients and increasing risk of emergency medical condition we are highly in the need of system where emergency situations can be tackled in a better and effective way .When in case the doctor is unavailable the patient may not risk his\her life .Thus ,we put forth this system through which we introduce this model .The goal of this approach is to recognize patients who are identical to index patient and derive insights from the records of identical patients to provide personalised predictions .It is to summarize & review and published studies describing computer-based approaches for predicting patients’ future health status based on health data & patient similarity , recognize gaps ,and provide a starting point for related future research. The model can be used as backend application for various hospital management systems in order to maintain the patients’ data which can be utilized in future .When emergency arises and the doctor in the worst situation is not available to attend the patient at very moment. This application will help to tackle emergency situation of a patient in an efficient way by taking reference from the patients’ records from the database.
2. LITERATURE REVIEW The major work in the area of patient-case similarity are: [1]
In this model the patient similarity method is applied for recognition of patient suffering from hepatocellular carcinoma who have gone Transarterial Chemoembolization.TACE is preferred only if some conditions are satisfied example no vascular invasion , etc. [2]
This model is used to summarize and review system oriented approaches for identifying patient health condition based on their data as similarity status providing a starting point for further research. [3]
This model explains about machine learning algorithms & statistical strategies that have been executed on big data mining. So as to make these algorithms more effective & avoid the problem of over fitting. Word2vec is a neural network that evaluates the text data .This model helps in understanding that word2vec is not only one algorithm instead includes two more learning models named as CONTINIOUS BAGS OF WORDS (CBOW) & Skipgram.
Page | 5 Research Publish Journals