Cancer Diagnosis Using Artificial Intelligence (AI) and Internet of Things (IoT)

Cancer Diagnosis Using Artificial Intelligence (AI) and Internet of Things (IoT)

DOI: 10.4018/978-1-6684-5422-0.ch004
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Cancer is an ailment that affects people from all walks of life. It is not age-specific, nor is it gender or race-specific. Affecting the cell cycle of various body parts like the brain, breast, etc., it increases the mortality rate, especially with the barriers in its early stage of detection. The advancement in technology has generated big datasets with high-resolution images. The oncologist's and clinician's diagnosis lacks accuracy, long time intervals, and limited information for advanced clinical care, influencing the survival rate. In the digital era, domain experts are reaping the importance of artificial intelligence (AI) techniques. As technology advances, AI and the internet of things (IoT) continue to escalate in the healthcare area, especially in cancer diagnosis. Researchers are looking for novel ways to diagnose cancer without the human-errors and false positives. Hence, the chapter focuses on all these imperative aspects of improved patient outcomes.
Chapter Preview
Top

Introduction

Cancer: An Overview

Cancer is an abnormality in cell proliferation i.e. not age, race, or gender-specific. It depends on the exposure to carcinogens like tobacco, UV radiation, etc., causing damage to the DNA (Barnes, 2018). In addition, there are certain lifestyle-associated carcinogens that directly or indirectly led to cancer. These include processed food, alcohol consumption combined with smoking, air pollutants (Anand, 2008), harmful chemicals released as a result of industrialization, etc. (Mitchell, 2021). Several studies have also shown that less than 10% of cancer is caused by genetic modifications (Edlich, 2005) whereas, the rest percentage is caused by carcinogenic agents, associated with environmental lifestyle factors. The cell-cycle damage (Barnes, 2018), failure in restoring the cell-repair mechanism (Kudravi, 2000), alteration in genes associated with cell growth (Hoenerhoff, 2022), and hormone therapy like breast enlargement (Rastelli, 2008) induce the development and advancement of cancer (Hoenerhoff, 2022). With the two major categories of benign and malignant, cancer can be classified depending on the parts of the body it is affecting. There are morphological differences between benign and malignant tumors. A benign tumor can become quite large, but it will not invade the nearby tissues or spread to other parts of the body. On the contrary, a malignant tumor can spread to other parts of the body, causing multiple organ failures, especially in the last stage. (Stavros, 1995) A diagrammatic representation illustrating its different types is shown in Figure 1. This showed various cancer types categorized with the subtypes and properties.

Figure 1.

Cancer overview with two major categories (i.e., benign and malignant) along with its various types

978-1-6684-5422-0.ch004.f01

The mode of treatment (Table 1) is dependent on the stage at which the cancer is been diagnosed. Including stage zero which is also called pre-cancerous, there are 5 stages leveling up from 0 to 4. Cancers that belong to stage zero are usually easy to treat and are considered to be a pre-cancer stage by most healthcare practitioners. Stage one and two tumors are specific to a particular area and can grow without affecting another cell. Till this stage, most of the cells can be benign. The malignancy occurs from stage three where the mass grows larger and coverage is till the lymph node and the surrounding tissues. Stage four is the last stage where the tumor outgrows and spreads through the entire body affecting several neighboring organs. This phenomenon of spreading and affecting is also called metastasis (Mustafa, 2016). The conventional methods consist of performing cytology on samples, scanning the affected area using X-Ray, CT, MRI, Sonography, etc., or undergoing Biopsy (Frangioni, 2008).

Table 1.
The common treatment with its action mechanism
TreatmentAction MechanismReference
Chemo-TherapyHigh power drugsChabner, 2005
Radiation-TherapyHigh dose of radiationSchaue, 2015
Hormone-TherapyHormonal doseDrãgãnescu, 2017
Immuno-TherapyInduce immune response Couzin-Frankel, 2013
Bone Marrow TransplantDamaged cells are replaced with healthy cellsCurtis, 1997

Sometimes with higher cancer coverage, the combination of treatments is implemented on the patients risking their survival rate. Moreover, these treatments have higher side effects like hair loss, appetite loss, nausea, fatigue ness, etc., and sometimes affect the psychology of the patient.

Key Terms in this Chapter

Semi-Supervised Learning: The algorithm is trained with incomplete data (labels or patterns) and the task is to identify output with the missing data, hence called as Semi-Supervised Learning.

Deep Learning (DL): It is a field of ML that uses a hidden neural network where two or more interrelated layers are present.

Unsupervised Learning: The algorithm is trained by certain data (labels or patterns) and the task is to relate the input to the output structure, hence called Unsupervised Learning.

Reinforcement Learning: The algorithm is trained at every step to learn and generate an accurate outcome, hence called Reinforcement Learning.

Supervised Learning: The algorithm is trained by certain data (labels or patterns) and the task is to identify the expected output which is manually supervised, hence called Supervised Learning.

Fuzzy Logic: It is a subset of DL which uses a vast spectrum of data and a heuristic plan of action, to reach an accurate conclusion (McNeill, 2014 AU257: The in-text citation "McNeill, 2014" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. ).

Convolutional Neural Network (CNN): The two major architectures of DL is Artificial Neural Network (ANN) a sub class of this is Convolutional Neural Network.

Internet of Medical Things (IoMT): is a subset of the Internet of Things (IoT) that describes the myriad of technologies developed and linked to exchange data via the internet and efficiently employed in the field of healthcare.

Complete Chapter List

Search this Book:
Reset