Design and Implementation Of A Web Based Diagnosis For Prostate Cancer

Abstract

Expert systems are now important tools in the medical field. They play a major role in early diagnosis before a patient meets a specialist. Diseases must be treated early to prevent complications. When they are not treated on time, they can lead to serious health problems and even death. These challenges are more severe in areas with few specialists, poor health facilities, and limited access to medical care.

To address these issues, this study designed and developed a web-based system that can diagnose and prescribe treatment for prostate cancer. The system uses a Machine Learning model, specifically Logistic Regression, to diagnose the disease and suggest treatment. It provides accurate results based on patient information. A dataset of 100 prostate cancer cases was used to train the model. The input attributes included Age of Patient (AP), Family History (FHPC), Painful Ejaculation (PE), Frequent Urination (FU), Blood in Semen (BS), Blood in Urine (BU), Weight Loss (WL), and Diet Intake. Prostate Risk (PR) served as the output. After training, the system achieved an accuracy of 95 percent.

CHAPTER ONE: GENERAL INTRODUCTION

1.1 Background of the Study

The rapid growth of internet technology has improved many human activities. It has also changed how computer systems support daily tasks (Omote et al., 2022). In the health sector, computer technology now supports diagnosis, surgery, and patient monitoring. Hospitals use tools such as surgical robots, wireless body area networks, and medical software. Some of these tools come from artificial intelligence, especially expert systems (Nohria, 2015).

An expert system stores the knowledge of a human expert and uses it to solve real problems. It provides quick and reliable support to users (Mottalib et al., 2016). These systems are useful in areas where medical experts are scarce. Many rural communities still face high rates of chronic diseases, disability, and death due to poor access to healthcare (Janssens et al., 2016). One of the most common conditions affecting men is prostate cancer, especially among those above forty years.

The prostate is a small gland in the male reproductive system. It produces fluid that supports sperm movement (Ellis, 2018). It sits under the bladder and surrounds part of the urethra. The gland has three main zones: central, peripheral, and transitional. Most prostate cancers develop in the peripheral zone.

Prostate cancer grows in the epithelial cells of the prostate. Its risk increases with age, and it is the second leading cause of cancer-related deaths in men after lung cancer (Bechis, 2011). It is common in Europe and the United States, and African American men face a higher risk (Folake et al., 2018).

In developing countries like Nigeria, poor health records make it difficult to track prostate cancer cases (Obansa et al., 2013). However, improved diagnostic tools such as MRI, CT scan, and PET scan have led to more reported cases. The disease now occurs at a rate of 9 cases per 100,000 people. It is the sixth most common tumor worldwide and the third most common among men. Due to its high prevalence, there is a strong need for fast, accurate, and low-cost diagnosis. This study, therefore, aims to design a web-based diagnosis and prescription system for prostate cancer.

1.2 Statement of the Problem

Prostate cancer remains the most common cancer among men. It continues to create social and economic burdens because it affects many people of African descent at a high rate (Delongchamps et al., 2017). Screening for the disease usually involves a Digital Rectal Examination (DRE) and a Prostate-Specific Antigen (PSA) blood test. Although DRE is cheap and widely used, it is inconsistent and often detects cancer only at an advanced stage (Naji et al., 2018).

Many patients also face long waiting times before diagnosis. In some cases, treatment arrives too late. These delays worsen the condition and may lead to death. Expert systems can help prevent such problems. An online diagnosis system provides early symptom detection, faster diagnosis, and quick recommendations. When its results match a doctor’s findings, it can save time and improve patient outcomes.

1.3 Aim and Objectives of the Study

The main aim of this study is to design and implement a web-based diagnosis system for prostate cancer.

The specific objectives are to:

• Review existing research on prostate cancer diagnosis and treatment.

• Collect prostate cancer data.

• Design a database to store patient information, medical history, expert knowledge, diagnosis results, and prescriptions.

• Build and train a Logistic Regression model for classifying prostate cancer cases.

• Test the system using real patient data.

1.4 Research Methodology

This project follows a systematic approach. The main steps include:

1. Reviewing literature, academic papers, and reports on online diagnosis systems for prostate cancer.

2. Collecting prostate cancer data from the University of Uyo Teaching Hospital to train and test the regression model.

3. Using MySQL as the backend database for storing patient information and the trained model.

4. Using Python and the Flask framework to develop the web-based diagnosis and prescription system.

1.5 Significance of the Study

This study benefits health organizations, government agencies, researchers, and the general public. The system reduces the workload of doctors by handling basic diagnosis and prescription tasks. It also lowers the risks that come from delayed diagnosis or self-medication.

Academically, the study adds to existing knowledge on prostate cancer diagnosis. It serves as a reference for students, lecturers, and future researchers.

1.6 Scope of the Study

This research focuses on the detection, diagnosis, and prescription of prostate cancer. The University of Uyo Teaching Hospital serves as the case study.

1.7 Limitations of the Study

The study faced several limitations.

Time was a major challenge. The project was completed under a short academic schedule, which limited the addition of some features.

Cost was another limitation. Limited funds restricted the availability of resources needed to fully implement the system.

Institutional policies also slowed down data collection. Some staff members were not willing to release key information due to security concerns. As a result, certain features could not be included.

1.8 Definition of Terms

Information System: A set of people, procedures, and equipment used to produce useful information.
Technology: The study and application of tools and processes used to achieve human goals.
Information: Data collected, stored, transmitted, and processed to convey a message.
Diagnosis: The process of identifying the cause of a disease through evaluation and laboratory tests.
CADx: Computer-Aided Diagnosis.
CADe: Computer-Aided Detection.
Prostate-Specific Antigen (PSA): A blood test used to screen for prostate cancer.
Symptoms: Signs that show the presence of a condition.
Vaccines: Measures taken to prevent diseases.