Article Preview
TopIntroduction
Dementia is a neurological disease that causes memory loss, as well as the inability to carry out daily tasks and maintain mental functioning. Although dementia is incurable, early detection can assist to alleviate symptoms. It refers to global impairment of intellect, memory, ability to comprehend and recognize people, loss of orientation and other cognitive disorder. It is among the leading causes of vulnerability among the elderly all over the world (Duff et al., 2008). According to the data of Global Dementia Observatory around 50 million cases of dementia is reported in which approximately 10 million new cases were reported on yearly basis. According to the survey, the overall number of people living with dementia is expected to reach 82 million by 2030 and almost 152 million in the year of 2050 (Lane et al., 2018). Among different types of dementia Alzheimer disease (AD) is the commonest form found and almost covered 60-70% of cases (Prince et al., 2016). First case of Alzheimer was reported in 1907 by Alois Alzheimer (Hippius & Neundörfer, 2003). AD is neurodegenerative disorder affects learning, cognitive behavioral performance and memory of the patient. Especially, the hippocampus region of the brain is responsible for learning and memory process. But due to deposition of amyloid beta (Aβ) protein or neurofibrillary tangles (NFT) causes AD (Hardy & Higgins, 1992).
Alzheimer's disease is a type of neurological disease that causes memory loss. Alzheimer is an irreversible neurodegenerative disease that affects and destroys our brain cells, brain cells of people. As a result, a person has reduced cognitive and functional ability emotional changes such as depression. Because all the diagnostic tools that are exists nowadays, they are either extremely expensive or time consuming or even invasive. Fig. 1 illustrates the brain's structure. The figure shows the mild cognitive impairment (MCI) to conversion to Alzheimer's disease. AD occurs when the neurons start dying due to the loss of glucose quantity.
Figure 1.
Progress of MCI to Severe AD (Shoukry et al., 2020)
Machine learning is a technique used to solve human problems without being explicitly programmed. Nowadays, machine learning has become very popular in almost many sectors such as banking, healthcare, social media, marketing and much more. Two popular learning techniques are used to solve the machine learning problem namely supervised learning and unsupervised learning (El-Sappagh et al., 2021). Recently these algorithms have become extremely popular in the healthcare domain. Above all, the medical field must deal with a bulk amount of data on a regular basis. As the volume increases, it becomes very difficult to handle huge data. So, we apply machine learning to them to make early predictions of disease. In the proposed paper, we explored machine learning principles including Adaptive boosting (AdaBoost), Extreme Gradient Boosting (XGBoost), Support vector machine, and Random Forest techniques for detecting Alzheimer's progression on the OASIS dataset.
This paper deals with a performance analysis of an intelligent dementia detection system built on machine learning techniques using the Oasis longitudinal dataset consisting of 150 subjects. The authors used various machine learning algorithms to train a dataset consisting of Alzheimer's disease sufferers or not. The author also compared them on a variety of performance criteria to demonstrate the machine learning algorithm's utility. Early detection of the disease can save a person's life. During the whole phases of Alzheimer's disease, a person may go through lots of changes not only in the brain but their walking and judgment skills lots of things are affected. This paper deals with the longitudinal dataset that consists of the following attributes that are mentioned below in the figure. The flow-graph of the suggested methodology is shown in Figure 2.
Figure 2.
Flow-graph of proposed approach