Artificial Intelligence is Taking on Parkinson's Disease
Michael Spencer
A.I. Writer, researcher and curator - full-time Newsletter publication manager.
Voice in Parkinson's Disease: A Machine Learning Study
AiSupremacy is all about covering the biggest stories I can find in artificial intelligence. You can also find me on iOS via a notification, if you prefer that to checking your Email.
If you enjoy articles about A.I. at the intersection of breaking news join AiSupremacy?here . I cannot continue to write without community support. (follow the link below).
https://aisupremacy.substack.com/subscribe
Read AI Supremacy in the new Substack app
Now available for iOS
Yesterday, I even lost followers when trying to cover?the topic of A.I. in warfare , but sometimes stories have to be told. While I’m not trained as a journalist, sometimes I cover angles that nobody else will.
As a writer, for every supporter who abandons you over not agreeing with your point of view, there may be another who shares the story because it?touches on something that traditional media doesn’t cover or loses in translation.?The untold stories. And there are many of them.
A.I. in Healthcare is Scaling in the 2020s
However, I also have a burning desire to tell more uplifting stories about A.I. in society. This is going to attempt to be such a story. Much of how A.I. is impacting our lives goes unnoticed in preliminary research that most people may be unaware of.
While I can find a summary of a study, I don’t often find an article to connect the dots and puts it in some perspective. Sometimes I will attempt that kind of synthesis, although I may not be fully qualified to do so.
This article is somewhat easier to read with Substack's formatting and images.
Parkinson’s Disease (PD) the Situation
According to ?the Parkinson’s Foundation, approximately?60,000 Americans?are diagnosed with Parkinson’s each year, and the disease affects more than ten million people worldwide.
Parkinson's disease is a brain disorder that leads to shaking, stiffness, and difficulty with walking, balance, and coordination. Parkinson's symptoms usually begin gradually and get worse over time. As the disease progresses, people may have difficulty walking and talking.
There is medication to treat the symptoms but it is notoriously difficult to prescribe exactly the right dose of Sinemet and other drugs which act to replace the missing dopamine that makes someone with Parkinson's shake or drag their foot.
Parkinson's disease (PD) is characterized by specific voice disorders collectively termed hypokinetic dysarthria.
Voice in Parkinson's Disease: A Machine Learning Study
Voice is abnormal in?early-stage?PD and as the disease progresses, voice increasingly degrades as demonstrated by high accuracy in the discrimination between healthy subjects and PD patients.
A recent study uses A.I. to help better define what stage of PD a patient is in.
PD patients frequently complain about variable impairment of voice emission.
In 2020, a team from Purdue University?leveraged? artificial intelligence? technology to collect and automatically measure the speech skills of people with Parkinson’s disease.
Jessica Huber, a?professor of speech, language, and hearing sciences and associate dean for research in Purdue’s?College of Health and Human Sciences , leads Purdue’s Motor Speech Lab. They developed innovative virtual studies to evaluate speech disorders related to Parkinson’s using artificial intelligence technology platforms even during the Covid-19 outbreak.
A year later in March, 2021 Researchers from the University of Florida said they?will use ?a $5 million grant from NIH to test a new artificial intelligence tool aimed at improving the diagnosis of Parkinson’s and related conditions.
The AI tool will?distinguish the precise diagnosis ?for early Parkinson’s disease or two related but distinct Parkinson’s-like syndromes. The three distinct neurodegenerative disorders – Parkinson’s disease, multiple system atrophy Parkinsonian variant (MSAp), and progressive supranuclear palsy (PSP) – can share overlapping motor and non-motor features, like changes in gait.
A.I. is very good at rapidly diagnosing conditions and identifying them better in novel ways than we were able to do before. Early detection and diagnosis can lead to a much better patient experience in healthcare.
领英推荐
The more recent study comes out of the University of Rome.
Machine Learning is Able to Classify Different Stages of PD
A.I. is also advancing how we understand things like voice changes in Parkinson’s disease. Machine learning methods have enabled the automatic classification on office impairment in various neurologic illnesses with high accuracy. However, only a few exploratory studies have been reported on the use of machine learning analysis in PD to date. It’s important to see if machine learning can distinguish between patients in different stages of the disease to see if it can recognize the effect of disease severity.?
The?new study by researchers in Italy and Jordan studied the voice of Parkinson’s disease patients in a large and clinically well-characterized cohort . This study is the first to classify voice in Parkinson’s disease patients based on the stage and severity of the disease and the effect of chronic L-Dopa medication. All diagnostic tests were evaluated for sensitivity, specificity, positive and negative predictive values, and accuracy.
Do you enjoy A.I. articles at the intersection of breaking news, then help me continue to write on the subject. I cannot continue to write without support. Grateful for all tips, patronage and community contributions.
Methods
The team clinically evaluated voices using specific subitems of the Unified Parkinson's Disease Rating Scale and the Voice Handicap Index. Voice samples recorded through a high-definition audio recorder under machine learning analysis based on the support vector machine classifier. We also calculated the receiver operating characteristic curves to examine the diagnostic accuracy of the analysis and assessed possible clinical-instrumental correlations.
Results
Voice is abnormal in?early-stage?PD and as the disease progresses, voice increasingly degradres as demonstrated by high accuracy in the discrimination between healthy subjects and PD patients in the?early-stage?and?mid-advanced-stage. Also, L-dopa therapy improves but not restore voice in PD as shown by high accuracy in the comparison between patients OFF and ON therapy. Finally, for the first time we achieved significant clinical-instrumental correlations by using a new score (LR value) calculated by machine learning.
Conclusion
A.I. is essentially able to create new biomarkers for PD’s development in its stages.
Voice is abnormal in?early-stage?PD, progressively degrades in?mid-advanced-stage?and can?be improved?but not restored by?L-Dopa. Lastly, machine learning allows tracking disease severity and quantifying the symptomatic effect of L-Dopa on voice parameters with previously unreported high accuracy, thus representing a potential new biomarker of PD.
Machine learning?is intersecting with the medical field and healthcare in more ways today than any one individual could possibly track. The team’s hypothesis is that machine learning analysis of speech samples is able to discriminate PD patients from controls, patients in?early?and?mid-advanced stages, and finally patients OFF and ON therapy, with previously unreported high accuracy. I find this pretty impressive and impactful that demonstrates that just in the last few years, A.I. is taking on PD like never before.
It’s possible that A.I. could be used even in the initial diagnosis of voice impairments and change how early we can diagnose P.D. Remember currently as it stands, there’s no specific test exists to diagnose Parkinson's disease.?Your doctor trained in nervous system conditions (neurologist) will diagnose Parkinson's disease based on your medical history, a review of your signs and symptoms, and a neurological and physical examination. A.I. could in this case with the biomarker accuracy, essentially speed up the early diagnoses in some cases.
A.I. is also boosting?Alzheimer’s disease classification? and predicting things like?A-Fib and stroke risk . These are all incredibly important and promising fields for further study. With each small success of A.I. in healthcare, it attracts more funding and researchers to the cause.
With A.I. on the job in healthcare, we won’t just be “augmented” by A.I. we will eventually literally be “cared for” by A.I. Machine learning eventually will help personalize the patient experience. The hospital or clinic of the future will be optimized by data and A.I at a level we may not realize today. Telehealth and the smart home will also have a lot more tools to keep us healthy and help us spot early signs with things are amiss.
Yet other studies have focused on how?wearable devices can help ?us monitor PD with the help of A.I.
As for the study from Rome, the researchers hope that their research will encourage the use of machine learning speech analysis for telemedicine techniques in Parkinson’s disease in the future.
Is A.I. the Holy Grail of Biomarker Tracking of Health?
Machine learning algorithms, which used a “likelihood ratio” of voice impairment ranging from zero to one, were also able to distinguish between “on” and “off” state patients.?The closer the likelihood ratio is to one, the greater the degree of voice impairment.
Paper: https://www.frontiersin.org/articles/10.3389/fneur.2022.831428/full
Reference: https://parkinsonsnewstoday.com/2022/03/08/machine-learning-identifies-patients-disease-stage-by-voice-changes/
What other biomarkers will A.I. be able to track that correlate to our underlying health? In the research there are many such examples. Often machine learning is able to pick up on patterns that humans would miss or be totally unable to notice.
Are Eyes Windows to the Soul of Health?
There’s some pretty whacky research of A.I. reading our retina or eyeballs to ascertain all kinds of things. Australian researchers have devised a new Artificial Intelligence (AI) programme that can help predict?a person's years of life?simply by looking at their retina—the tissue at the back of the eye.?Predicting mortality risk? at a glance? You have to sort of see it to believe it.
Age-related macular degeneration (AMD) is the biggest cause of sight loss in the UK and USA and is the?third largest cause of blindness ?across the globe. A.I. could also predict sight-threatening conditions. The research from May 2020, that was a collaboration between Google Health, DeepMind and Moorfields Eye Hospital was?published in Nature Medicine .?It showed that artificial intelligence (AI) has the potential to not only?spot the presence ?of AMD in scans, but also predict the disease’s progression.?
A.I.’s Impact on Clinical Diagnosis, Tracking and Treatment Will Only Expand
Clearly A.I. can be implicated in diagnosis and disease tracking in a way that could revolutionize healthcare. When applied it will be able to improve how our medical interventions are admitted.
We must fund more research at the intersection of healthcare, machine learning and science. Better data will enable us to tailor care better to the patient. This personalization is essentially to improve healthcare outcomes and reduce the cost of healthcare simultaneously.
When a biomarker is found, everything in our understanding of data can change. Such is the case I think with voice alterations and PD. In the Rome study 84% of the patients included in their cohort (97 out of 115 patients) manifested a variable degree of clinically overt voice impairment.
Also, they found a clinically overt voice impairment in 68% of?early-stage?patients and 100% of?mid-advanced-stage?patients. Thus it’s a pretty reliable indicator to track the stages of PD. This can help clinicians to time treatments better and also reduce patient anxiety by having a clearer diagnosis. This scenario can be generalized to many health conditions and chronic diseases, where A.I. can contribute in their management.
If you enjoy articles about A.I. at the intersection of breaking news join AiSupremacy?here . I cannot continue to write without community support. (follow the link below).
https://aisupremacy.substack.com/subscribe
PE modernizing industrial safety and power grid; MS, innovation/design strategist; inventor, entrepreneur
2 年Torben Riise
Very interesting Michael, this could provide a really great advance for preventing the later stages of those that could suffer by catching it early. ??
Author of What "Do I DO" appeared in Forbes Magazine, Media including Fox and NBC, Radio and talk shows.
2 年Very interesting
Building Connections
2 年This is amazing :)