The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma

Oral squamous cell carcinoma (OSCC) is the sixth most frequent cancer worldwide. It is responsible for 80-90% of all mouth malignant neoplasms and has a mortality rate of up to 50%. Oral cancer has multifactorial etiology, mainly smoking, tobacco, alcohol consumption, betel quid chewing, and high...

Full description

Saved in:
Bibliographic Details
Main Author: Azeem, Muhammad
Format: Thesis
Language:English
Published: 2023
Subjects:
Online Access:http://eprints.usm.my/59894/1/MUHAMMAD%20AZEEM%20-FINAL%20THESIS%20PSGD001119%28R%29-E.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
id my-usm-ep.59894
record_format uketd_dc
spelling my-usm-ep.598942024-02-15T03:40:28Z The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma 2023-07 Azeem, Muhammad R Medicine RC254-282 Neoplasms. Tumors. Oncology (including Cancer) Oral squamous cell carcinoma (OSCC) is the sixth most frequent cancer worldwide. It is responsible for 80-90% of all mouth malignant neoplasms and has a mortality rate of up to 50%. Oral cancer has multifactorial etiology, mainly smoking, tobacco, alcohol consumption, betel quid chewing, and high-risk human papillomavirus (HPV). A total of 57 patients were recruited from the clinic at the Hospital Universiti Sains Malaysia (USM). In this retrospective study, advanced computational statistical modeling techniques were used to evaluate data descriptions on three risk factors studies: sociodemographic, clinicopathological, and surgical margins features of OSCC patients. The R-Studio software and syntax were used to design and develop the hybrid biometry method, implement, and the odd ratio. The advanced approach was executed in three parts, such as developing syntax for R for the biometry hybrid method which consists of data bootstrap methodology, multiple layer feedforward neural network (MLFFNN), and logistic regression. Male gender, smoking, betel quid, and alcohol habit variables were significantly related to death (p < 0.05). Among clinicopathological features increasing tumour size, metastasis, moderately and poorly differentiated OSCC, and Ki67 expression were significantly related to deceased patients (p < 0.05). Furthermore, features of surgical margins perineural invasion, bone invasion, and involvement of surgical margins were significantly related to the death of OSCC patients (p < 0.05). This finding might contribute to the underlying cause of poor prognosis. In conclusion, there exist potential risk factors in relation to OSCC in the Malaysian population. The conclusion of the study might illustrate the superiority of the hybrid model technique used in the study. 2023-07 Thesis http://eprints.usm.my/59894/ http://eprints.usm.my/59894/1/MUHAMMAD%20AZEEM%20-FINAL%20THESIS%20PSGD001119%28R%29-E.pdf application/pdf en public phd doctoral Universiti Sains Malaysia Pusat Pengajian Sains Pergigian
institution Universiti Sains Malaysia
collection USM Institutional Repository
language English
topic R Medicine
R Medicine
spellingShingle R Medicine
R Medicine
Azeem, Muhammad
The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma
description Oral squamous cell carcinoma (OSCC) is the sixth most frequent cancer worldwide. It is responsible for 80-90% of all mouth malignant neoplasms and has a mortality rate of up to 50%. Oral cancer has multifactorial etiology, mainly smoking, tobacco, alcohol consumption, betel quid chewing, and high-risk human papillomavirus (HPV). A total of 57 patients were recruited from the clinic at the Hospital Universiti Sains Malaysia (USM). In this retrospective study, advanced computational statistical modeling techniques were used to evaluate data descriptions on three risk factors studies: sociodemographic, clinicopathological, and surgical margins features of OSCC patients. The R-Studio software and syntax were used to design and develop the hybrid biometry method, implement, and the odd ratio. The advanced approach was executed in three parts, such as developing syntax for R for the biometry hybrid method which consists of data bootstrap methodology, multiple layer feedforward neural network (MLFFNN), and logistic regression. Male gender, smoking, betel quid, and alcohol habit variables were significantly related to death (p < 0.05). Among clinicopathological features increasing tumour size, metastasis, moderately and poorly differentiated OSCC, and Ki67 expression were significantly related to deceased patients (p < 0.05). Furthermore, features of surgical margins perineural invasion, bone invasion, and involvement of surgical margins were significantly related to the death of OSCC patients (p < 0.05). This finding might contribute to the underlying cause of poor prognosis. In conclusion, there exist potential risk factors in relation to OSCC in the Malaysian population. The conclusion of the study might illustrate the superiority of the hybrid model technique used in the study.
format Thesis
qualification_name Doctor of Philosophy (PhD.)
qualification_level Doctorate
author Azeem, Muhammad
author_facet Azeem, Muhammad
author_sort Azeem, Muhammad
title The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma
title_short The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma
title_full The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma
title_fullStr The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma
title_full_unstemmed The development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma
title_sort development of a new prognostic logistic regression model: a case study on oral squamous cell carcinoma
granting_institution Universiti Sains Malaysia
granting_department Pusat Pengajian Sains Pergigian
publishDate 2023
url http://eprints.usm.my/59894/1/MUHAMMAD%20AZEEM%20-FINAL%20THESIS%20PSGD001119%28R%29-E.pdf
_version_ 1794024067997630464