As the fastest-growing neurological condition in the world, never has the need to find an effective treatment for Parkinson’s disease been more pressing.
With no known cure, those already living with the disease are all too aware of the timescales involved in drug discovery and development.
Aside from the average timescale from lab to clinical approval taking around 10–15 years at a cost of over $1–2 billion, studies also indicate that up to 90% of clinical drug development programs fail.
However, artificial intelligence (AI) may help tip the scales as scientists implement AI to pioneer new ways of identifying drug compounds capable of tackling this debilitating neural disease — yet in a fraction of the time of traditional research methods.
In a new study published in Nature Chemical Biology, Professor Michele Vendruscolo and his team at Cambridge University found that machine learning could speed up the initial screening process tenfold and ‘reduce the cost by a thousandfold’.
With what appears to be an almighty leap in the right direction, we’ll delve into how AI facilitates the fight against Parkinson’s disease, from its ability to help diagnose and treat to being a key component in a successful drug discovery.
Key Takeaways
- Parkinson’s affects over 8.5 million people worldwide, with numbers rising faster than any other neurological disease.
- A Cambridge University study demonstrates AI’s ability to accelerate drug discovery and development timescales and reduce costs in the fight against the disease.
- Additional studies show machine learning algorithms provide methods of early detection and diagnosis in sufferers.
- Meanwhile, AI applications help to reduce the impact of debilitating symptoms associated with the disease.
The Need for Speed in Parkinson’s Disease Treatment
With over 8.5 million people globally living with Parkinson’s, its devastating symptoms are all too visible in those who are battling with the disease.
However, potentially, the most alarming element of the illness is no one is immune.
Prominent figures, including Muhammad Ali, Michael J Fox, and Billy Connolly, have all publicly battled Parkinson’s disease, raising awareness of how the disease — whose symptoms cause unrelenting tremors, balance and mobility issues, and muscle stiffness — affects sufferers.
Professor Michele Vendruscolo’s research was carried out at the Yusuf Hamied Department of Chemistry at the University of Cambridge, focusing on how AI-based strategies could speed up research significantly.
AI’s Role in Drug Discovery
“One route to search for potential treatments for Parkinson’s requires the identification of small molecules that can inhibit the aggregation of alpha-synuclein, which is a protein closely associated with the disease,” said Professor Vendruscolo.
“This is an extremely time-consuming process – just identifying a lead candidate for further testing can take months or even years.”
While certainly not limited to searching for treatment medications for other diseases, including cancer, the study focused on identifying chemical compounds that could block the accumulation of the alpha-synuclein protein.
Here, machine learning algorithms were able to screen a chemical library of millions of compounds, rapidly leading the team to identify five highly potent compounds for further investigation. A task that, through traditional research, may have taken months, if not years, to complete.
Parkinson’s Diagnoses and Aftercare AI-Powered Innovations
Identifying treatments that could prevent or tackle the onset of Parkinson’s is still in the research stage, and there is clearly still a long way to go.
Despite the potential acceleration of the research and drug development phases, should a preventive element be discovered, researchers will still need to complete lengthy regulatory and clinical trials before a widespread rollout of treatments occurs.
However, in the meantime, AI is also advancing the medical profession’s ability to detect symptoms early and treat Parkinson’s patients.
Fascinating studies across the scientific world are investigating ways to detect the early onset of this infamously challenging disease to diagnose.
An example of this is the Massachusetts Institute of Technology (MIT) AI-centric study, which can detect the ‘presence and severity’ of Parkinson’s disease from a person’s breathing patterns.
Monitoring a sample of 7,687 people, including 757 Parkinson’s patients, MIT researchers created a series of neural network algorithms mimicking how a human brain works. Remarkably, their results were able to determine if an individual has Parkinson’s from their breathing as they slept.
Even once diagnosed, people living with Parkinson’s are also now benefiting from the advancements in AI-led treatments.
For example, backed by Harvard Innovation Labs, care firm CareYaya is developing algorithms to create personalized programs such as AI art therapy sessions for patients with Parkinson’s.
CareYaya says their AI application ‘uses any tablet or touchscreen, using tactile motions, which are then turned into images that allow for visual discovery, memory prompts, and creative expression’ in patients.
Additionally, SpeechVive has created a wearable AI device that improves the speech clarity of people coping with Parkinson’s. The innovative gadget creates a “babble” sound cue whenever a patient is talking, which helps trigger them to speak clearer and louder.
The continual advancement of artificial intelligence is likely to fuel further breakthroughs in the fight against Parkinson’s disease, along with many other illnesses and diseases. Whether it’s pioneering early detection methods or providing personalized symptom management and treatments. Ultimately, cures may emerge from using AI to process data at a speed unmatched by humans.
It is analogous to when Google, DeepMind, and AlphaMissense AI classified 89% of the 71 million potential ‘errors’ in DNA — a stark contrast to the mere 0.1% that had taken human experts years to explore.
The Bottom Line
If AI can speed up the research and development phase while also mitigating the cost of drug development and regulation, the medical field has a powerful new tool. Harnessing the power of machine learning algorithms for drug and treatment research won’t just benefit patients who fall victim to this cruel disease.
For now, every innovative study utilizing AI edges scientists and doctors closer to winning battles on various fronts.