Assignment Task

Introduction

1.1 Research Background and Motivation

Technological development is considered one of the major factors in recent times as it helps to improve the quality of life and resolve the issues and challenges faced in daily life. The main importance of technological development in daily life is that it helps resolve the daily challenges faced by society and provides a quality and happy life (Hacker et al., 2020). In recent times, happy life has been considered one of the main requirements for people as most people live under stress and face several mental diseases like anxiety, depression, Bi-polar, obsessive-compulsive disorder, schizophrenia, Psychosis, and stress etc. Among them, depression is major disorder, which is faced by mostly young generation. According to the world health organization (WHO), 5% of adults suffers from depression, which is approximately 280 million people in the world. Every year 7 million people die due to depression and this number is growing every year. In addition, most of patients are not diagnosed and treated as 75% of patients are from low- or middle-income countries.

Clinical diagnosis of depression is another major challenge for patients as well as doctors. Firstly, Current diagnosis is symptom-based process in which doctors counsel the patient and identify symptoms of depression like hopelessness, mood swings, change in body weight, uncontrollable emotions, feeling low, insomnia, suicidal tendency, and anxiety. Based on the severity of symptoms, doctors usually provide treatment to the patients. However, this diagnosis process has several disadvantages. Firstly, it is time consuming process as doctors need hours of counselling. Majorly, in the underdeveloped countries where governments face problem of shortage of doctors, and doctors are unable to spend enough time to the patients for counseling, thus misdiagnosis may be possible. Secondly, once patient is diagnosed with depression and treatment is started, monitoring the treatment efficacy also depends on the counselling as there is no proven clinical diagnosis method available for the monitoring of depression treatment. Thirdly, diagnosis majorly depends on the doctor’s skill of counselling. Sometimes misdiagnosis may be possible while counselling by the unskilled doctors. Lastly, patients are sometimes not aware about the symptoms of depression, which may cause misdiagnosis.

Because of these challenges in the diagnosis of disorders, US Food and Drug Administration (FDA) encouraged and approved several diagnosis tools, which use AI (Artificial intelligence) and machine learning. AI/ML based diagnosis tools are approved by US FDA for diseases related to cancer, cardiovascular, ophthalmology, radiology, endocrinology, neurology and hematology. However, there is no single AI/ML based approved diagnosis method available in recent times, which can diagnose mental disorder like depression.

Therefore, computer aided solutions are required which could effectively diagnose depression with high accuracy, less time and lower cost. The computer aided diagnosis method is required to monitor treatment progress.

1.2. EEG signal an effective biomarker for depression.

For any computer aided diagnosis solution, effective biological marker is required for diagnosis of disease. For mental disorders, monitoring EEG signals may be used for monitoring the neural activities (signals) which elicit from brain as EEG signals change at different mental stages (Mahatab Rohh-Azizi et al.). Thus, EEG can be explored as an effective biological marker and diagnosis tool for computer aided diagnosis solution for depression. However, EEG signals are much complex and messy in nature, it is very difficult to differentiate EEG signals of a patient with depression and a healthy person with naked eye. Thus, in recent times, many ML and DL (Deep learning) tools have been explored using EEG signal for detection of mental disorder like Alzheimer, Epilepsy, Parkinson’s, Seizure, and Schizophrenia. Thus, this research work utilized EEG signals as a biomarker and proposes computer aided detection system for diagnosis of depression. 

1.3. Computer aided detection system for depression

Machine Leaning (ML) is widely used in smart phones, computers and robotics and well-known industry like healthcare. Many researchers have explored the promise of machine learning based disease diagnosis (MLBDD), which is inexpensive and time efficient. Deep learning (DL) is one type of machine learning based on the artificial neural network in which multi-layer of processing are used to identify progressively higher-level features from the data. Deep Neural Networks sometimes referred to as neural networks with many layers, are utilized in the field of DL to evaluate and model complex data (Gourisaria et al., 2020). These networks can learn and identify patterns from vast amounts of data, including images, text, and voice. Applications of deep learning include speech and pattern identification, natural language processing and self-driving cars. In the medical sector, the main use of deep learning is that it helps in several diagnosis by identifying patterns in the biomarkers (as input raw data) which normally use images, graphs and signals in the diagnosis process. This capability of identifying patterns in the input raw data can be used for diagnosis of mental disorder like depression (Chang Su et al. 202).

Different mechanisms and techniques have been used in deep learning for the diagnosis of mental disorders. The main techniques used in deep learning include attention mechanism, Convolutional Neural Networks (CNN), Personalized Medical Treatments, Natural language processing and long short-term memory (LSTM) (Jang et al., 2020). The main role of the attention mechanism is that it improves the interpretability of the service provided by a medical firm to its customers and enhances performance in various aspects such as accuracy of diagnosis, effectiveness of treatment plans, and overall patient satisfaction. The attention mechanism is a machine learning (ML) technique that enables a model to concentrate on inputs while processing them. Attention mechanisms have a range of wide uses in the medical field, including machine image analysis, Natural Language Processing (NLP), and drug inspection.

Convolutional Neural Networks (CNNs) are a type of deep learning algorithm that is used particularly for image classification tasks (Sadeghi et al., 2019). There is a range of applications where they can be used including medical imaging, to detect and diagnose various mental disorders such as schizophrenia, depression and Alzheimer`s disease. One of the key benefits of CNN is to automatically learn features from images, which can be used to classify images into different categories (Chen et al., 2019). This is particularly useful in medical imaging, where there may be a large amount of variability in the appearance of a given disorder. CNN`s can be trained on a large dataset of images and then used to classify new images, which can help improve diagnoses` accuracy and consistency. In addition to CNN, personalized medical treatments also play an important role in the treatment of mental disorders. Personalized medicine involves tailoring treatment plans to the individual patient, considering their genetic makeup, medical history, and other factors. This can help improve treatment effectiveness and reduce the risk of side effects. Natural language processing (NLP) is also playing an important role in the field of mental health. NLP techniques can be used to analyse amounts of large text data, such as posts of social media, to identify patterns and trends that may be related to mental health (Uban et al., 2021). For example, NLP can be used to analyse tweets to identify individuals at risk of developing depression. NLP can also be used to analyse the content of therapy sessions and to identify patterns of behaviour that may be related to mental disorders. The use of CNNs, personalized medical treatments, and NLP are helping to improve diagnosis, treatment, and management of mental disorder. These techniques are helping to improve the accuracy and consistency of diagnosis while also providing new insights into the underlying causes of mental disorders. As these technologies continue to advance, it is assumed to be that they will play an increasingly crucial role in the field of mental health.

The CNN model has the capability to stride over signal data in different hidden layers, but output is not fed back to the network. Thus, CNN models are good in mining the features of EEG signal but are poor in learning information in a sequence. Thus, additional layers of models are required which have capability to process sequential information. Recurrent Neural Network (RNN) is suitable for modelling short term sequential memory but not useful for long term memory. However, LSTM (long short-term memory) is a type of recurrent neural network (RNN) (Tulensalo J at al. 2020). LSTM model was developed to resolve the problem of long-term dependence and vanishing gradient problem. This model contains memory cells and gates to control network information and remember this information for long period of time. Thus, CNN and LSTM hybrid model is required for diagnosis of depression.

1.4. Problem Domain

Mental disorder is considered one of the major issues people have recently faced. As per the report of WHO the ratio of people that are facing issues with mental disorders is 1 out of every 8 people throughout the world or it can be said that there were around 970 million people that are facing the issue of mental disorders in the year 2019 (WHO, 2022). Several consequences of mental disorder issues include anxiety, depression, bipolar disorder, Post-Traumatic Stress Disorder, Eating Disorders and Schizophrenia (Kakhramonovich, 2022). Among them depression is a major mental disorder which affects mostly young and adults. However, diagnosis of depression is critical due to over dependence on doctor’s skill, time and patient clarity on symptoms. In present time, there is no FDA approved diagnosis method available for depression. Thus, AI/ML based computer aided solution is required which can help the screening and treatment monitoring for depression patient.

1.5. Aim and objectives.

Deep learning is considered a major technology in recent times as it helps deal with unstructured data and organize it to structured data. This research aims to explore the potential of deep learning in analysing electroencephalogram (EEG) data for the diagnosis of depression patients. The main objectives of this research include:

• To identify the major symptoms and challenges faced by patients with depression.
• To examine the role of deep learning into the diagnosis of depression
• To evaluate the effectiveness of deep learning in analysing EEG data for depression diagnosis

1.6. Research Questions

The main research questions to be addressed include:

1. What are the challenges faced by doctors while screening or diagnosis of depression?

2. How can deep learning contribute to resolve the challenges related to diagnosis of depression?

3. What is the potential of deep learning in analyzing EEG data for the diagnosis of depression?

4. What is the contribution of hybrid neural networks for EEG based diagnosis of depression?

1.7. Significance

In recent times mental disorders have been considered one of the major issues that people face. The major impact of mental disorder is that it leads to the generation of several issues such as anxiety, depression, bipolar disorder, Post-Traumatic Stress Disorder, Eating Disorders and Schizophrenia. Among them, depression is one of the major issues people face with a mental disorder. The major impact of depression is that it leads to initially decrease patient self-confident and loss of interest in their life and ultimately ends in suicidal thoughts and death. An excessive amount of depression also leads to the loss of mental thinking ability and leads to the death of a person. It is necessary to mitigate the impact of depression specifically in young people so that youngsters can live freely. To reduce mental disorders, it is important to first diagnose depression in patients. This will reduce the incidence of misdiagnosis, reduce burden on the doctors and improve patient lifestyle. As per discussed by Byrd and Lipton (2019), the major role of DL is that it helps minimize the risk by improving the accuracy and speed of diagnosis, which can lead to earlier intervention and treatment. Additionally, DL can help identify patterns and relationship in EEG data that may be difficult for human clinicians to detect, leading to more personalized and effective treatment plans for patients with mental disorders.

In addition, Kwekha-Rashid et al. (2021) also discussed the importance of DL in the medical industry and described that in recent time world is full of diseases and outbreaks, which is impacting human life. To improve the quality of humans and provide an effective lifestyle, deep learning is considered one of the major technologies that helps to assist in tasks such as detection, diagnosis and treatment planning by monitoring efficacy of treatment.

The main significance of this research is that it helps to discuss the techniques and methods that can be used in the detection of mental disorders and provide efficient and timely diagnosis that can be treated effectively. This research explores deep learning in the diagnosis of depression using EEF signals. The main importance of deep learning is that it can process a large number of datasets and deal with the unstructured data present in the EEG signal dataset. The major deep learning techniques used in the medical sector include Convolutional Neural Networks (CNN) and long short-term memory (LSTM).

1.8. Methodology

The primary objective of the thesis is to develop an efficient novel hybrid model (CNN+LSTM) for the diagnosis of depression. As shown in Figure 1.1, the diagnosis of depression mainly contains four steps: Raw EEG data collection, EEG data pre-processing, hybrid model (CNN+LSTM) EEG data processing and model diagnosis.

1.9. Thesis Organization

Chapter 1 is about the introduction of the research. This chapter helps to discuss the background of the entire research along with the description of the main aim and objectives on which the entire work is based. This chapter also helps to describe the significance of the entire work. Chapter 2 reviews the literature related to techniques for detecting mental disorders. Chapter 3 presents the EEG data and the methods used to diagnose depression. Chapter 4 presents the results and discusses the findings. Chapter 5 concludes the research and describes the main findings and discusses future work.