ME/CFS Research Chatbot

Dolphin

Dolphin

Senior Member (Voting Rights)
(copied from: https://forums.phoenixrising.me/threads/me-cfs-research-chatbot.93488/ )
Introducing the ME/CFS Research Chatbot! It's meant to provide a reliable, accurate, and up-to-date summary of published research for the general public.

While general LLMs like ChatGPT and Claude might provide fine answers to queries about myalgic encephalomyelitis/chronic fatigue syndrome, the sources are often unreported or of unknown quality, and the chatbots are prone to hallucination and to change answers from one moment to the next.

The chatbot that I have designed is a bit different: It relies only on research abstracts from the NIH's PubMed database that address this disease, it can provide a list of the abstracts consulted, and it's designed to provide the same, 'best' answer each time a particular question is asked - including admitting ignorance, if information is lacking.

For the technically inclined: The chatbot is a RAG system using Gemini 2.0 Flash to interpret the 250 abstracts that best match a user's query. The abstracts are embedded in a vector database using Chroma. And the frontend was built with Gradio and resides on Hugging Face. The full dataset of abstracts, with metadata, is here.

If you try it, please let me know how it goes.
 
Looks like a cool idea, but unfortunately think it has a major GIGO (garbage in, garbage out) problem. Ie. It’s referencing a bunch of studies that aren’t exactly reliable and known to inflate their claims and such and such in their abstract. So it tends to repeat the popular “memes” in the research.

See
IMG_1504.jpeg
Click to expand...


But perhaps if used critically to maybe look at the current literature, without assuming the literature is accurate, it may be useful?
 
I asked it if exercise is safe for ME/CFS. It gave some vague contradictory answers that both said it might be beneficial for some, and that NiCE doesn’t recommend GET. Including this:
  • Potential Benefits: Exercise therapy can have a positive effect on fatigue, physical functioning, and sleep in adults with ME/CFS, compared to usual care or passive therapies. Some studies even suggest it might be more effective than adaptive pacing.
Click to expand...
I think this is a prime example of the garbage in, garbage out that Yann mentioned.

Edit: it was also essential to work with a professional to get a positive result!
 
I’ve messed around with some edge cases, so to speak:

Question:
Does your personality make it more likely that you can get ME/CFS?

It appears that certain personality traits may make you more likely to develop ME/CFS. However, it's important to note that research in this area has had mixed results, and more studies are needed to fully understand the connection.

Here's what the research suggests:
  • Specific Traits: Some studies suggest that individuals with ME/CFS may have a tendency to be perfectionistic, conscientious, hardworking, neurotic, and introverted. They may also have high personal standards and a strong desire to be socially accepted.
  • Neuroticism and Harm Avoidance: Several studies have found that higher levels of neuroticism (a tendency to experience negative emotions) and harm avoidance are associated with ME/CFS.
  • Lower Extraversion and Novelty Seeking: Some research indicates that people with ME/CFS may be less extraverted and have lower novelty-seeking tendencies.
  • Perfectionism: Qualitative evidence suggests that perfectionism may play a role in the development and continuation of ME/CFS. Studies have found that people with ME/CFS may report higher levels of perfectionism, particularly concerning doubts about actions and concern over mistakes, as well as lower self-esteem.
  • Self-Directedness and Cooperativeness: Lower self-directedness (responsibility, purposefulness, resourcefulness, and self-acceptance) and cooperativeness (tolerance, empathy, helpfulness, and compassion) have also been observed in ME/CFS patients, especially those with co-occurring psychiatric conditions.
It's important to remember that these are just tendencies observed in some studies, and not everyone with these traits will develop ME/CFS. Also, some studies have found no major differences in personality between ME/CFS patients and healthy controls.

Disclaimer: This information is for general knowledge and informational purposes only, and does not constitute medical advice. It is essential to consult with a qualified healthcare professional for any health concerns or before making any decisions related to your health or treatment.

Question:
Should ME/CFS patients face their fears?

While I cannot provide medical advice, I can share information from the provided research about whether ME/CFS patients should face their fears.

One study from 2019 indicated that stronger fear avoidance beliefs were associated with higher fatigue and worse social adjustment in young people with severe ME/CFS. The authors of this study suggested that fearful beliefs about activity could be targeted using cognitive-behavioral interventions.

Please remember that this information is for educational purposes only and does not substitute professional medical advice.

Question:
Is Lightning Process good for people with ME/CFS?

The Lightning Process (LP) is a 3-day training program that combines concepts from Neuro-Linguistic Programming, Life Coaching, and Osteopathy. Experiences of young people who have undergone the Lightning Process to treat chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) have been reported as mostly positive.

One study showed that seven out of nine young people reported satisfaction and much improvement after undergoing the Lightning Process, while two reported dissatisfaction and no improvement.

Helpful aspects included the theoretical rationale, practical exercises, and the technique learned. Less helpful aspects were the intensity and short duration of the treatment with little follow-up, the secrecy surrounding it, and feelings of being blamed if the treatment did not work.

Another study showed that physical function was better in those allocated specialist medical care (SMC) plus the LP, and this improved further at 12 months. At 6 months, fatigue and anxiety were reduced, and at 12 months, fatigue, anxiety, depression, and school attendance had improved in the SMC+LP arm.

The LP is effective and is probably cost-effective when provided in addition to SMC for mild/moderately affected adolescents with CFS/ME.

Disclaimer: This information should not be considered medical advice. Please consult with a qualified healthcare professional for personalized guidance.
 
Utsikt said:
Question:
Is Lightning Process good for people with ME/CFS?
Click to expand...
Google AI is much better
The Lightning Process is not recommended for people with ME/CFS by major health organizations. NICE (National Institute for Health and Care Excellence) in the UK explicitly states not to offer it. The World ME Alliance and its members also do not endorse it. While some individuals may report positive experiences with the Lightning Process, these are not consistently supported by evidence and may be due to factors other than the treatment itself.

Here's why it's not recommended:
  • Lack of robust evidence:
    There's a lack of strong, peer-reviewed evidence to support the Lightning Process as an effective treatment for ME/CFS.
  • Potential for harm:
    Some people with ME/CFS may experience negative consequences from the Lightning Process, such as increased fatigue or emotional distress.


  • Emphasis on unproven claims:
    The Lightning Process, which combines neurolinguistic programming and osteopathy, claims to cure ME/CFS in a short period, which is not supported by scientific evidence.


  • NICE guidelines:
    NICE guidelines strongly advise against the use of the Lightning Process for ME/CFS, emphasizing the need for evidence-based treatments
 
It is not just the GIGO problem, AI thus far has not shown any great capacity to independently reason, including understanding methodology and context.

At this point it is little more than a fancy collator and summariser. Still does not in any sense understand the actual content of the material it is working with. It has not gone meta yet.

Though, in fairness, you could say all that about some human experts as well. More so in some fields than in others.
 
You must log in or register to reply here.
Back
Top Bottom