Early Detection of Dementia using AI

Our technical challenge was to get as much information as possible out of non-invasive imaging methods such as MRI scans, and it’s not the information that you can get from a single scan but also how the brain changes over time is important.
— Professor Wiro Niessen, Scientific Director of Quantib

According to the Alzheimer’s Society, there are around 80,000 people in the U.K. with dementia. One in three people over 65 will develop dementia, and two-thirds of people with dementia are women. The Society predict that this number will increase to one million by 2021 and 1.7 million by 2051.

 

The earliest signs of dementia are usually memory problems, confusion, and changes in the way a person behaves and communicates. Cognitive symptoms of dementia can include like problem solving, difficulty learning new skills and impaired decision making. Behaviour changes can include fear, insecurity, anger and often, depression like symptoms.

Alzheimer’s and Frontotemporal Dementia

The University College London’s (UCL) Institute of Healthcare Engineering was able to utilise machine learning to develop an algorithm that can diagnose with clarity the disorder a patient is suffering from. This algorithm is known as SuStaIn (Subtype and Stage Interference). Analysts at UCL fed the machine those patients MRI scans who are dealing with dementia. Now, it can display medical professionals the precise location of parts of the brain where neurodegeneration (death of neutrons) is occurring and classify the distinct subgroups of both Alzheimer and Frontotemporal dementia.

Vascular Dementia

Dementia symptoms: Vascular dementia is the second most common type after Alzheimer's disease. Credit: Getty Images

Dementia symptoms: Vascular dementia is the second most common type after Alzheimer's disease. Credit: Getty Images

Lewy Body Dementia (LBD), is estimated to effect approximately 1.4 million Americans. LBD includes patients from both dementia with Lewy bodies and Parkinson’s disease dementia diagnoses. People with LBD often experience the same symptoms as Alzheimer’s patients, in addition to sleep disturbances, visual hallucinations, muscle rigidity or other Parkinsonism physical and cognitive impairments.

Lewy Body dementia

Lewy Body Dementia (LBD), is estimated to effect approximately 1.4 million Americans. LBD includes patients from both dementia with Lewy bodies and Parkinson’s disease dementia diagnoses. People with LBD often experience the same symptoms as Alzheimer’s patients, in addition to sleep disturbances, visual hallucinations, muscle rigidity or other Parkinsonism physical and cognitive impairments.

Lewy Body Dementia (LBD), is estimated to effect approximately 1.4 million Americans. LBD includes patients from both dementia with Lewy bodies and Parkinson’s disease dementia diagnoses. People with LBD often experience the same symptoms as Alzheimer’s patients, in addition to sleep disturbances, visual hallucinations, muscle rigidity or other Parkinsonism physical and cognitive impairments.

Lewy body dementia and Alzheimer’s are often mistaken for each other due to the similarity of their symptoms. AI can assist with this issue as it is able to identify the precise areas of the brain where damage has occurred as well as identify the cause. AI can distinguish between these two disorders by identifying the amount of shrinkage caused to the hippocampus using machine learning. A lack of shrinkage in the hippocampus is a sign of Lewy body Dementia rather than Alzheimer’s.

Parkinson’s Disease

These kinds of diseases takes place when nerve cells are unable or lacks to produce brain chemical called ‘dopamine’. Neurons are the functional units of brain; when they are unable to response then such disorders can also occur. It can be due to nervous weakness, blockage of veins, diabetes disorders and many more.

These kinds of diseases takes place when nerve cells are unable or lacks to produce brain chemical called ‘dopamine’. Neurons are the functional units of brain; when they are unable to response then such disorders can also occur. It can be due to nervous weakness, blockage of veins, diabetes disorders and many more.

Tencent (a Chinese technology company) and Medicaid (a medical firm established in the U.K.) formed a partnership to create AI when monitoring patients with Parkinson’s disease. They collaborated in developing an app that produces a video test for find hand movements. The app uses the device’s camera to record the patient opening and closing their hands. The recording is then turned into a graph which is then sent to a medical professional for examination.

 

Currently, the app takes 30 minutes to send results however Tencent and Medopad are working to decrease the hold-up time to 3 minutes. In order to do so, the AI is being prepared using deep learning to score the test results by itself, completely bypassing the step of sending the footage to a medical professional. As of right now, the app is in its testing stages. John Hopkins University is utilizing the app to treat their patients and soon this AI will be accessible for all clinical and public health projects.

Huntington’s Disease

Analysis of a patient’s speech is the modern strategy of tracking the progression of Huntington’s disease. However, this method is neither cost effective nor time efficient so, researchers at the University of Michigan Medicine led by Professor Emily Mower have created AI that can resolve the issue. The software can analyse and record a patient’s voice and detect Huntington’s disease with 81 percent accuracy. The software can also analyse how the symptoms of Huntington’s develop over time by using deep learning and provide perceptions to medical professional of the best treatment methods for the disorder.

AI is changing the way we view neurodegenerative disorders and the brain in general. We are moving toward a future where neurodegenerative disorders can be diagnosed with 100% accuracy and find definitive treatments.

 

Article thumbnail credit: Artificial Intelligence for Dementia Diagnosis – Genetic Analysis, Speech Analysis, and More, [Click here for page]. Note that this article has not contributed to any of the written content of my article, and that you click the above link at your own discretion as the page has not been checked by our team.