- Shrinidhi R
1) What is the impact of this research in the next 4 years?
Ans: Apart from Alzheimer’s Disease, many other neurodegenerative or neuropsychiatric disorders like autism and schizophrenia don't have an accurate diagnostic measure. Finding the causes, treatments and preventions of such diseases has been the focus of most of the research fields.
ADiag: Graph Neural Network Based Diagnosis of Alzheimer’s Disease, helps in accurate diagnosis of Alzheimer's disease based on neural thickness and connections using an MRI scan. It has been found that weak connections and abnormal thickness has also been associated with autism and other neurodegenerative disorders.
Thus, this method would be an effective diagnostic measure not only for Alzheimer’s Disease, but also for the other neurodegenerative disorders, with accurate modifications. Thus, it would have a great impact in the field of psychiatry and clinical psychology in the next four years.
2) How does the research paper connect to another field of study?
Ans: Conversion of neural connections in the brain into an electrical circuit form has been a part of the physics research field for many years now. This research paper connects mathematics, physics, biology and psychology, and thus is an interdisciplinary research project.
The connections in the human brain can be figured out after brain autopsy, and the inferences can be used for the research. Brain autopsy has also been related to research in neural specialisation and anatomy. Apart from Alzheimer’s Disease, many other neurodegenerative or neuropsychiatric disorders like autism and schizophrenia etc don't have an accurate diagnostic measure, thus this research paper is also connected to the research that is being performed in other neurodegenerative disorders related fields.
Graynet Processing and FreeSurfer Processing, which was used for processing the scans in the research is an open access, industry standard MRI analysis software, has been a part of biotechnology, computer science and biomedical engineering research fields.
3) What is the Application of this research in our lives?
Ans: Apart from Alzheimer’s Disease, many other neurodegenerative or neuropsychiatric disorders like autism and schizophrenia etc don't have an accurate diagnostic measure. Currently, only qualitative means of testing are employed in the form of scoring performance on a battery of cognitive tests.
The inherent disadvantage of this method is that the burden of an accurate diagnosis falls on the clinician’s competence.
Quantitative methods like MRI scan assessment are inaccurate at best, due to the elusive nature of visually observable changes in the brain. Apart from MRI scans, there exists tests like The Mini Mental State Examination (MMSE) which requires clinicians to evaluate potential patients on a 30-point test on attention, memory, language, orientation and visual-spatial skills.
Another is the Clinical Dementia Rating (CDR), which is a 5-point test on a similar rubric, but also includes home affairs and personal care. Irrespective of the specifics, a correct diagnosis is solely based on the clinician’s competence and not on quantitative backing; this is the major cause of the high rate of misdiagnosis.
Thus, ADiag is superior in each and every respect compared to extant AD diagnostic methods due to its non-invasive nature, use of commonly available T1 weighted MRI scans, superior accuracy, high scalability and robust construction. It would improve the quality of life of the neurodegenerative disorders diagnosed patients, due to effective, accurate and fast diagnosis of the disorder. The rate of misdiagnosis would also reduce.
4) How has the current research materials and methods helped this research?
Ans: The current physics research findings regarding the conversion of neural connections into an electrical circuit has played an important role in the ideation of this research idea.
Graynet Processing and FreeSurfer Processing, which was used for processing the scans in the research is an open access, industry standard MRI analysis software, has been a part of biotechnology, computer science and biomedical engineering research field’s current findings.
Brain biopsy and brain autopsy, a neuro-oncological method has also been useful in this research. Psychological testing methods such as The Mini Mental State Examination (MMSE) and Clinical Dementia Rating (CDR), has been used in this research.
5) How do you think the research will be perceived by the general public? Summarize the paper in a manner that is easy for a person from non-science background to understand?
Ans: This research will be perceived as a method of diagnosing neurological disorders like Alzheimer’s and schizophrenia used by psychologists and psychiatrists, by the general public. Alzheimer’s Disease (AD) is the most widespread neurodegenerative disease, affecting over 50 million people across the world. While its progression cannot be stopped, early and accurate diagnostic testing can drastically improve quality of life in patients.
ADiag, is a method of diagnosing Alzheimer’s disease by plotting the graph of the neural connections and neural networks. The neural network of the normal human brain is generalized after performing autopsy of more than 60 brain samples, after the patient's death. This data is used for plotting the graph of the normal brain connections. Similarly, the neural network connections of Alzheimer’s disease patients are collected through biopsy and autopsy.
During the process of diagnosis, the MRI scans are then processed on FreeSurfer, and the thickness data extracted by FreeSurfer was then processed by the graph generation software Graynet. The data obtained from these two processes are used for graph plotting. The graph is plotted through software devices, and the graph obtained is used for diagnosis purposes, after comparing it with the sample graphs.
6) As a future researcher, what do you consider the shortcomings of this research?
Ans: The sample data used for human brain network analysis was only 60, thus the accuracy of the graph plotted might not be reliable. Since, human brain is complicated, generalization of neural connections is not accurate, as it might vary with patients, thus leading to misdiagnosis. Quantitative methods like MRI scan assessment are inaccurate at best, due to the elusive nature of visually observable changes in the brain.
Since MRI scans were used for network generalization, the conclusions may not be accurate. Thus, the accuracy of the test is only 83%. Combined with an expansion in dataset size from 121 to 300 graphs and a revamped architecture, ADiag’s accuracy values can be increased. This method also uses expensive devices, thus might not be affordable to the general public.
For more , visit - https://arxiv.org/pdf/2101.02870.pdf?
Comments