1. Sign our petition calling on Cochrane to withdraw their review of Exercise Therapy for 'CFS'.
    Dismiss Notice
  2. Guest, the 'News in Brief' for the week beginning 24th June 2024 is here.
    Dismiss Notice
  3. Welcome! Read the Core Purpose and Values of our forum.
    Dismiss Notice

Unsupervised cluster analysis reveals distinct subtypes of ME/CFS patients based on peak oxygen consumption and SF-36 scores, 2023, Lacasa et al

Discussion in 'ME/CFS research' started by Andy, Oct 7, 2023.

  1. Andy

    Andy Committee Member

    Messages:
    22,309
    Location:
    Hampshire, UK
    HIGHLIGHTS
    • ME/CFS is a disabling chronic disease with a lack of diagnostic tests.
    • Oxygen consumption is a possible biomarker of CFS.
    • O2 consumption allows classifying patients status according to the Weber's classification.
    • A worse Weber's classification infers a worse outcome on the SF-36 questionnaire.
    • Unsupervised machine learning is a powerful tool for analyzing data.
    Abstract

    Purpose
    Myalgic encephalomyelitis, commonly referred to as chronic fatigue syndrome (ME/CFS), is a severe, disabling chronic disease and an objective assessment of prognosis is crucial to evaluate the efficacy of future drugs. Attempts are ongoing to find a biomarker to objectively assess the health status of (ME/CFS), patients. This study therefore aims to demonstrate that oxygen consumption is a biomarker of ME/CFS provides a method to classify patients diagnosed with ME/CFS based on their responses to the Short Form-36 (SF-36) questionnaire, which can predict oxygen consumption using cardiopulmonary exercise testing (CPET).

    Methods
    Two datasets were used in the study. The first contained SF-36 responses from 2,347 validated records of ME/CFS diagnosed participants, and an unsupervised machine learning model was developed to cluster the data. The second dataset was used as a validation set and included the cardiopulmonary exercise test (CPET) results of 239 participants diagnosed with ME/CFS. Participants from this dataset were grouped by peak oxygen consumption according to Weber's classification. The SF-36 questionnaire was correctly completed by only 92 patients, who were clustered using the machine learning model. Two categorical variables were then entered into a contingency table: the cluster with values {0,1} and Weber classification {A, B, C, D} were assigned. Finally, the Chi-square test of independence was used to assess the statistical significance of the relationship between the two parameters.

    Findings
    The results indicate that the Weber classification is directly linked to the score on the SF-36 questionnaire. Furthermore, the 36-response matrix in the machine learning model was shown to give more reliable results than the subscale matrix (p − value < 0.05) for classifying patients with ME/CFS.

    Implications
    Low oxygen consumption on CPET can be considered a biomarker in patients with ME/CFS. Our analysis showed a close relationship between the cluster based on their SF-36 questionnaire score and the Weber classification, which was based on peak oxygen consumption during CPET. The dataset for the training model comprised raw responses from the SF-36 questionnaire, which is proven to better preserve the original information, thus improving the quality of the model.

    Open access, https://www.sciencedirect.com/science/article/pii/S0149291823003521
     
    Andy, Oct 7, 2023
    #1
    ukxmrv, Ash, Hoopoe and 1 other person like this.
  2. Hutan

    Hutan Moderator Staff Member

    Messages:
    27,829
    Location:
    Aotearoa New Zealand
    I haven't read this yet, but is peak oxygen consumption just VO2 max? In which case, it's hardly surprising that low VO2max goes with low SF-36 scores. It could just be that people with low SF-36 scores tend to be more unfit than people with high scores.

    Also, the authors don't seem to be considering the unusual variability of VO2 max in people with ME/CFS. If you put someone in a category based on their VO2 max on one day, but on another day they have a much lower VO2max, where does that leave the categorisation?
     
    Hutan, Oct 7, 2023
    #2
    ukxmrv, alktipping, Ash and 6 others like this.
  3. inox

    inox Senior Member (Voting Rights)

    Messages:
    539
    Location:
    Norway

     
    inox, Oct 7, 2023
    #3
    alktipping, Ash, Trish and 3 others like this.
  4. Trish

    Trish Moderator Staff Member

    Messages:
    53,405
    Location:
    UK
    They have used machine analysis of SF-36 scores across all the SF-36 fields, physical, social and emotional to divide pwME into two groups. What I can't see anywhere in the paper is any data on which factors contributed to being in each group.

    I'm not sure studies like this that use very complicated machine learning on questionnaire data actually take us further forward, especially when they don't show which factors in the SF-36 range of questions contributed to people being allocated to two groups.

    Given that other studies tend to show pwME score lower on physical and social functioning and close to normal on emotional factors, I'm guessing the main difference between the groups is simply a division according to severity ME/CFS.
     
    Trish, Oct 7, 2023
    #4
    ukxmrv, EndME, Sean and 3 others like this.

Share This Page