Between 17 and 24 million individuals globally suffer from chronic fatigue syndrome, a debilitating condition that poses significant challenges in terms of identification. The World Health Organization defines this issue, also known as myalgic encephalomyelitis, by a diverse array of symptoms that converge to produce a debilitating and difficult-to-explain feeling of persistent, chronic stress. These symptoms include insomnia, post-exertional malaise, and cognitive impairment, often leaving individuals struggling to carry out daily tasks and potentially bedridden.
Marcos Lacasa, a researcher pursuing his PhD in Bioinformatics at the Universitat Oberta de Catalunya (UOC), highlights the diagnostic complexities surrounding this condition, emphasizing the absence of definitive tests or biomarkers for precise patient characterization. He asserts that the diagnostic process heavily relies on the physician’s expertise and the patient’s medical history. Early intervention plays a crucial role in shaping the disease’s progression.
In his recent publication in the open-access journal Scientific Reports by Nature, Lacasa delves into the application of machine learning, a form of artificial intelligence, to enhance disease comprehension and diagnostic accuracy. Collaborating with Jordi Casas from UOC’s Applied Data Science Lab, José Alegre from the Vall d’Hebron Institute of Research (VHIR), and fellow researcher Ferran Prados, Lacasa explores innovative approaches to address the diagnostic challenges associated with chronic fatigue syndrome.
Presently, the absence of definitive tests renders the diagnosis of chronic fatigue syndrome reliant on subjective assessments through questionnaires evaluating individuals’ perceived fatigue levels. Although research efforts advocate for objective measures such as oxygen consumption assessments, standardized surveys like the 36-Item Short Form Health Survey (SF-36) remain prevalent in clinical practice. However, early diagnosis and treatment initiation continue to present hurdles in managing this condition effectively.
Lacasa proposes leveraging machine learning algorithms to generate patient profiles based on questionnaire responses, streamlining the identification of myalgic encephalomyelitis-related symptoms and expediting patient referrals to specialized care units. By synthesizing data from standard questionnaires, the concept of “synthetic patients” emerges, enabling the emulation of real patient data for research purposes without compromising individual privacy.
While this innovative approach offers potential benefits for academic research, Lacasa underscores the necessity of authentic input data like that obtained from the SF-36 questionnaire to maintain the integrity of the analysis. Beyond questionnaire-based diagnostics, ongoing research endeavors aim to identify genetic markers for precise diagnostic tools and potential therapeutic targets. Despite the absence of a definitive cure, current treatments focus on symptom management through lifestyle modifications, therapies, and targeted medications.
Lacasa advocates for increased funding to facilitate genomic sequencing in individuals with myalgic encephalomyelitis, emphasizing the potential of genetic analyses in identifying therapeutic targets. By elucidating the role of specific proteins in the disease, researchers aim to develop tailored interventions that address the core symptoms effectively, paving the way for improved management strategies in the future.
