GPT for ME/CFS Questions

Yann04

Senior Member (Voting Rights)
Initial posts have been moved from another thread.

Generative AI turns out to be only somewhat useful for now, but in the near future it will be very good at it. So there will be opportunities to do more of this.

I just noticed that now GPT-models pre trained on data are available for free users too. Is someone up to the task of training one on Me-pedia, trial by error, and science for ME (if members consent). It would be so much more useful for us than vanilla chatgpt. I’m only half joking with the proposal.

edit: forestglip knows more about this than me, please refer to the message below mine.
 
Last edited by a moderator:
I just noticed that now GPT-models pre trained on data are available for free users too. Is someone up to the task of training one on Me-pedia, trial by error, and science for ME (if members consent). It would be so much more useful for us than vanilla chatgpt. I’m only half joking with the proposal.

By pre-trained, do you mean fine tuned, or are you talking about the custom GPTs, which is basically just giving it default prompts and attachments. The former would be ideal as you can give it unlimited data, and I've thought a lot about that possibility, but it's still very expensive, as far as I know.

The second is good if the amount of data isn't too big. And it returns much more accurate answers. But all of MEpedia might be too much.
 
I just noticed that now GPT-models pre trained on data are available for free users too. Is someone up to the task of training one on Me-pedia, trial by error, and science for ME (if members consent). It would be so much more useful for us than vanilla chatgpt. I’m only half joking with the proposal.

edit: forestglip knows more about this than me, please refer to the message below mine.
It's not the only option, but OpenAI has a system to train specialist models. I don't know what it involves but it should be possible to do something like this soon enough. So the idea already exists, I just don't know how much work would be involved and how much it would cost.

And if it's complicated for now, one rule with AI models is that whatever limitations and complications is involved in using them is likely to be solved not much later than the time it'd take to work around those. So it's best to wait until it's easy to do it. Which kind of works for us, as if it's too much work, we can't do it anyway.

The next big thing is like to be what's called agents. It mostly amounts to being more autonomous, capable of solving intermediary steps and doing more than simply responding to prompts. So at this point a capable agent could simply be given the task of building a specialist model using the best resources out there.
 
(note - earlier posts in this thread have been moved from the linked thread)

Continuing on with the conversation from another thread, I set up a custom GPT, named MEGPT for now, to be able to give it some default relevant documents so that it can answer questions about ME/CFS effectively.

Accessible at this link, though you'll need a ChatGPT account: https://chatgpt.com/g/g-AvMmt4r3f-megpt

So far it only has the Mayo Clinic Proceedings Diagnosis and Management webpage.

I'll put all the files that it has access to on Google Drive: https://drive.google.com/drive/folders/1FaBFOiUYZCXmmnM1RcS8lcvqyg0rfDaE?usp=drive_link

The data limit is actually larger than I had thought, because it's using something called embeddings, which means it stores a bunch of chunks of the documents and then only searches for and uses the relevant parts for answering questions. It says up to 2 million tokens per file, and up to 20 files. 2 million tokens is somewhere around 1.5 million words. So it might actually be able to use all of MEpedia. Though I'm not quite sure how I'd turn it into text files or if I'd even be allowed to.


Here are the custom instructions I gave it:
You are a helpful assistant who provides useful, detailed information about ME/CFS and related diseases like Long COVID. Only use information from the attached files, and any answers using the files should include any relevant citations if they are provided in the documents. If the question can not be answered from the files, do not answer.

Although it seems to be ignoring the "include citations" part.

Testing difference in answers between this GPT and regular ChatGPT:

"What treatments exist for ME/CFS?"

MEGPT:
The treatment of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) involves a multi-faceted approach, as there is no single cure. The primary focus is on managing symptoms and improving quality of life. Here are the main strategies:

### Symptom Management

1. **Post-Exertional Malaise (PEM):**
- **Pacing/Rest:** Managing activity levels to avoid triggering PEM.
- **Stimulus Reduction:** Minimizing sensory overload.

2. **Fatigue:**
- **Pacing:** As with PEM, balancing activity and rest.
- **Medications:** Low-dose naltrexone and low-dose aripiprazole may be beneficial.
- **Diet and Supplements:** Anti-inflammatory diets, treating vitamin deficiencies.

3. **Sleep Issues:**
- **Medications:** Melatonin, trazodone, suvorexant, doxepin, gabapentin, and pregabalin.

4. **Cognitive Dysfunction:**
- **Aids:** Journaling, memory aids, occupational therapy.
- **Medications:** Low-dose naltrexone and low-dose aripiprazole, with careful use of stimulants.

5. **Orthostatic Intolerance:**
- **Fluids and Electrolytes:** Increased intake.
- **Medications:** Fludrocortisone, midodrine, propranolol, pyridostigmine, guanfacine.

6. **Dizziness:**
- **Therapies:** Vestibular therapy.
- **Medications:** Low-dose SSRIs or SNRIs.

7. **Muscle or Joint Pain:**
- **Medications:** Over-the-counter pain relievers, duloxetine, milnacipran, pregabalin, gabapentin, low-dose naltrexone.

8. **Neuropathy:**
- **Medications:** Pregabalin, gabapentin, tricyclic antidepressants.
- **Therapies:** Compression or brace therapy.

9. **Sensory Amplification:**
- **Tools:** Noise-canceling headphones, tinted glasses.
- **Medications:** Low-dose aripiprazole.

10. **Gastrointestinal Symptoms:**
- **Diet:** Anti-inflammatory diets, small meals.
- **Medications:** Pro/synbiotics, antidiarrheals, antihistamines, fiber, motility agents.

### Monitoring and Support

- **Symptom and Activity Diary:** Keeping track of symptoms and activities to identify patterns and triggers.
- **Wearable Devices:** Monitoring physical activity and heart rate to prevent PEM.

### Pharmacologic Treatments

- **Low-Dose Medications:** Pyridostigmine, aripiprazole, and naltrexone.
- **Stimulants:** Modafinil or methylphenidate for severe fatigue and brain fog (used cautiously).

### Supplements

- **Metabolic and Cognitive Support:** Coenzyme Q10, nicotinamide adenine dinucleotide (NADH), L-arginine, omega-3 fatty acids (EPA, DHA).
- **Immunologic and Anti-Inflammatory Support:** Lactoferrin, quercetin, curcumin.
- **Vitamin and Mineral Supplements:** To address deficiencies.

### Specialty Care

- **Referral to ME/CFS Specialists:** For complex cases that require more intensive management.
- **Intravenous Therapies:** Saline and electrolyte repletion in severe cases.

### Comprehensive Care

- **Total Care:** For patients with severe ME/CFS, including parenteral nutrition and specialized home or nursing facility care.

Overall, the treatment plan should be tailored to the individual, starting with low doses of medications and adjusting based on tolerance and response. The focus is on improving functional capacity and quality of life while avoiding exacerbation of symptoms【5†source】.

ChatGPT:
Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is a complex and poorly understood condition, and there is currently no cure. Treatment focuses on managing symptoms and improving quality of life. Here are some common approaches:

1. **Lifestyle Changes:**
- **Pacing:** Balancing activity and rest to avoid exacerbating symptoms.
- **Sleep Management:** Developing good sleep hygiene practices.

2. **Medications:**
- **Pain Relievers:** To manage muscle and joint pain.
- **Antidepressants:** Sometimes prescribed to help with sleep and pain.
- **Sleep Aids:** For those with sleep disturbances.
- **Low-dose Naltrexone (LDN):** Some evidence suggests it may help reduce symptoms.

3. **Therapies:**
- **Cognitive Behavioral Therapy (CBT):** Can help patients cope with the condition, though its efficacy is debated.
- **Graded Exercise Therapy (GET):** A controversial treatment that involves gradually increasing physical activity. Many patients and advocacy groups have raised concerns about its safety and efficacy.
- **Physical Therapy:** Tailored exercises to help improve mobility and reduce pain.

4. **Dietary Adjustments:**
- **Nutritional Supplements:** Some patients find vitamins and minerals like B12, magnesium, and coenzyme Q10 helpful.
- **Balanced Diet:** Maintaining a healthy diet to support overall well-being.

5. **Alternative Therapies:**
- **Acupuncture:** Some patients report relief of symptoms.
- **Massage Therapy:** Can help reduce pain and improve relaxation.

6. **Psychological Support:**
- **Counseling:** To help manage the emotional impact of living with a chronic illness.
- **Support Groups:** Connecting with others who have ME/CFS can provide emotional support and practical advice.

It's important for patients to work with healthcare providers to develop an individualized treatment plan. Research is ongoing to better understand ME/CFS and to find more effective treatments.
 
Last edited by a moderator:
Though it's free for others to use, I had to make a Plus account for $20 per month to create it. I'm not a big fan of paying OpenAI, and I would prefer to do this with Anthropic Claude, as they seem to care more about AI safety, but as yet, Claude doesn't have anything similar to custom GPTs, but hopefully in the future.
 
Okay, this is really cool. Instead of downloading all of MEpedia, which would have its advantages, but would be pretty difficult, I wouldn't get updates, and I'm not sure I'm allowed, I instead made the GPT able to access the live MEpedia API. So basically it can browse the wiki to answer questions. When you ask it a question, first it uses the search to find the most relevant pages, then it goes to those pages to get the answer.

New instructions:
You are a helpful assistant who provides useful, detailed information about ME/CFS and related diseases like Long COVID. You have access to files with information, and you can access pages on ME-Pedia, a wiki for information about ME/CFS. First search the wiki for relevant pages, then access the most relevant result(s) to get the information needed to answer the question. Only use information from the files or the wiki.

Example conversation:
User: Who founded ME Action?

MEGPT: ME Action was co-founded by Jennifer Brea and Beth Mazur on May 12, 2015.


User: What is dextro naltrexone?

MEGPT: Dextro-naltrexone is a potential treatment proposed for chronic pain, fatigue, and cognitive disorders related to brain inflammation. It is the right-handed stereoisomer of the drug naltrexone. Unlike the left-handed form (levo-naltrexone) used in low-dose naltrexone (LDN) therapy, dextro-naltrexone is not expected to have opioid-blocking effects. This characteristic potentially allows it to be dosed higher than LDN, providing greater anti-inflammatory and symptom relief benefits without the harsh side effects associated with opioid receptor antagonism.


User: what was the incline village outbreak?

MEGPT: The Incline Village outbreak, which occurred in the winter of 1985 in Incline Village, Nevada, involved an estimated 160 residents who became ill with what was then thought to be chronic Epstein-Barr virus but is now recognized as chronic fatigue syndrome (CFS) or myalgic encephalomyelitis (ME). Local doctors, including Dr. Paul Cheney and Dr. Daniel Peterson, treated many of the affected individuals. The outbreak gained attention due to the high number of cases and the severe, flu-like symptoms that left many residents debilitated.

A similar outbreak occurred around the same time at a high school in Truckee, California, where nine out of ten teachers using a specific conference room experienced symptoms typical of sick building syndrome and CFS.

The Centers for Disease Control and Prevention (CDC) eventually investigated the outbreak, but their response was criticized for not examining the patients directly and dismissing the disease as affecting mainly "depressed menopausal women." This dismissive attitude led the media to dub the disease "Yuppie Flu".

The potential causes of the outbreak have been speculated to include reactivation of herpes viruses, such as Epstein-Barr virus and human herpesvirus 6 (HHV-6), due to immunologic dysfunction. Another theory proposed by patient and mold avoidance pioneer Erik Johnson suggested that a combination of biotoxins, including toxic mold and cyanobacteria from Lake Tahoe, might have made residents more susceptible to viral infections.

For the three responses, here were the searches and pages it automatically used to get the information:

1.
Search: "ME Action founder"
Read: "The MEAction Network"

2.
Search: "dextro naltrexone"
Read: "Dextro-naltrexone"

3.
Search: "Incline Village outbreak"
Read: "1984 Incline Village chronic fatigue syndrome outbreak"

Honestly, kind of scared at how well AI already works and how easily I could put it together.

Edit: It looks to not be able to get pages that are too long, such as the Pace Trial page. Which lol even has a note on the page that it needs to be shortened.
 
Last edited:
@forestglip this is so cool!

I gave your model a badly written article about ME, the kind you often see on health websites, and asked it to make bullet points detailing errors, it was pretty well done, to the point that with some fine tuning and proofreading it would be emailable to article creators.

### Errors in the Article about ME/CFS

- **Definition of CFS/ME**:
- The article implies that severe and long-lasting fatigue is called Chronic Fatigue Syndrome (CFS). However, ME/CFS is a complex, chronic disease characterized by more than just fatigue. It includes a range of symptoms like post-exertional malaise, unrefreshing sleep, cognitive impairment, and orthostatic intolerance [oai_citation:1,Mayo-Clinic-Proceedings---Diagnosis-and-Management-of-Myalgic-Encephalomyelitis_Chronic-Fatigue-Syndrome.docx](file-service://file-j31jL7bPmrqJZVtOaKgR7w4x).

- **Prevalence Rates**:
- Dr. Priti Shanker's statement that ME/CFS affects 10-15% of middle-aged women is incorrect. The actual prevalence of ME/CFS in the general population is about 0.2-0.4% [oai_citation:2,Mayo-Clinic-Proceedings---Diagnosis-and-Management-of-Myalgic-Encephalomyelitis_Chronic-Fatigue-Syndrome.docx](file-service://file-j31jL7bPmrqJZVtOaKgR7w4x) [oai_citation:3,Mayo-Clinic-Proceedings---Diagnosis-and-Management-of-Myalgic-Encephalomyelitis_Chronic-Fatigue-Syndrome.docx](file-service://file-j31jL7bPmrqJZVtOaKgR7w4x).

- **Characterization of CFS/ME Symptoms**:
- The article describes CFS as "unexplainable exhaustion for over six months without any cause," but ME/CFS is more accurately characterized by the combination of severe fatigue, post-exertional malaise, unrefreshing sleep, and other symptoms persisting for at least six months [oai_citation:4,Mayo-Clinic-Proceedings---Diagnosis-and-Management-of-Myalgic-Encephalomyelitis_Chronic-Fatigue-Syndrome.docx](file-service://file-j31jL7bPmrqJZVtOaKgR7w4x).

- **Diagnosis Criteria**:
- The article should emphasize that ME/CFS diagnosis is not just based on fatigue lasting over six months, but also on the presence of post-exertional malaise, unrefreshing sleep, and either cognitive impairment or orthostatic intolerance, as per the 2015 Institute of Medicine (IOM) criteria [oai_citation:5,Mayo-Clinic-Proceedings---Diagnosis-and-Management-of-Myalgic-Encephalomyelitis_Chronic-Fatigue-Syndrome.docx](file-service://file-j31jL7bPmrqJZVtOaKgR7w4x).

- **Exercise and Treatment Recommendations**:
- The recommendation for gentle exercise routines, like short outdoor jogs, is misleading. Exercise can worsen symptoms in ME/CFS patients due to post-exertional malaise. Instead, pacing and energy management strategies are recommended [oai_citation:6,Mayo-Clinic-Proceedings---Diagnosis-and-Management-of-Myalgic-Encephalomyelitis_Chronic-Fatigue-Syndrome.docx](file-service://file-j31jL7bPmrqJZVtOaKgR7w4x).

- **Management of ME/CFS**:
- The article's focus on self-care and relaxation techniques oversimplifies the management of ME/CFS. While these can be helpful, a more comprehensive approach that includes pacing, pharmacological management, and treating comorbid conditions is necessary [oai_citation:7,Mayo-Clinic-Proceedings---Diagnosis-and-Management-of-Myalgic-Encephalomyelitis_Chronic-Fatigue-Syndrome.docx](file-service://file-j31jL7bPmrqJZVtOaKgR7w4x).

For more accurate and detailed information on ME/CFS, you can refer to the Mayo Clinic Proceedings and the CDC guidelines:
- https://me-pedia.org/wiki/Mayo_Clinic_Proceedings
- https://me-pedia.org/wiki/Chronic_fatigue_syndrome_diagnostic_criteria

Feel free to reach out for further clarification or additional information.
 
Okay, this is really cool. Instead of downloading all of MEpedia, which would have its advantages, but would be pretty difficult, I wouldn't get updates, and I'm not sure I'm allowed, I instead made the GPT able to access the live MEpedia API. So basically it can browse the wiki to answer questions. When you ask it a question, first it uses the search to find the most relevant pages, then it goes to those pages to get the answer.

I haven't the faintest idea how any of this works, but can you also ask it to prioritise other fairly reliable sources? Like the 2021 NICE guideline?

If so, maybe it could even learn about the Larun review problem... :laugh:
 
I’m wondering if you could either add this document for the chatgpt model: https://www.s4me.info/threads/the-name-question.332/ and/or ask it in instructions to prioritise the naming ME/CFS (or ME)

If a moderator could chime in to let me know if I can use posts from the library, then sure. For now, I added the Wikipedia page for ME/CFS which talks about naming, and added a bit in the instructions to use ME/CFS over CFS.

I haven't the faintest idea how any of this works, but can you also ask it to prioritise other fairly reliable sources? Like the 2021 NICE guideline?

The three files it has now are Mayo Clinic Proceedings, the Wikipedia page, and the NICE guidelines, plus access to all of MEpedia.

Current instructions:
You are a helpful assistant who provides useful, detailed information about ME/CFS and related diseases like Long COVID. You have access to files with information, and you can access pages on ME-Pedia, a wiki for information about ME/CFS. First search the wiki for relevant pages, then access the most relevant result(s) to get the information needed to answer the question. Only use information from the files or the wiki.

Instead of CFS, refer to the disease as ME/CFS.

Link to relevant pages at end of answer. When generating links, ensure that all URLs are absolute. For example, instead of using /wiki/Fibromyalgia, use https://me-pedia.org/wiki/Fibromyalgia. Do not create a title for these links and instead list the entire url.

Links if necessary:
2021 ME/CFS NICE Guidelines: https://www.nice.org.uk/guidance/ng206

If so, maybe it could even learn about the Larun review problem... :laugh:
Hmm not sure what this is.
 
Last edited:

Oh yeah, very similar! Looks like she put a lot of work into it. I think since then the GPT Actions have been released which allows direct access to other websites' APIs (like MEpedia) which makes things a lot easier than downloading lots of MEpedia, which it looks like she was doing. I haven't looked at all the posts in detail, but there's probably some good ideas there to add to this.

Also the file limit has been increased by a lot since then.
 
Looks like there's the issue of many MEpedia articles being out of date. I wonder if MEAction could hire people to write for the wiki. I think there are paid writers for certain topics on Wikipedia. (Paid by third parties, not by Wikimedia.)

"please tell me about decodeme timelines"
ME/GPT: The DecodeME project is a significant UK-based study aimed at understanding the genetic factors involved in ME/CFS by analyzing DNA profiles of individuals with ME/CFS compared to healthy controls. Here is an overview of the timeline for the DecodeME study:

1. **July 2020**: The DecodeME website was launched, allowing patients to sign up for updates and provide contact details. Recruitment for the study was initially planned to start in March 2021 .

2. **August 2020**: The official start of the study was expected in early 2021 .

3. **November 2020**: The London School of Hygiene and Tropical Medicine announced it would no longer be involved in the DecodeME project, and key investigators from LSHTM also stepped down .

4. **June 14, 2021**: Ethics approval was granted for the study, with the REC reference 21/NW/0169 .

5. **November 2021**: It was announced that the recruitment had been delayed until 2022, starting with a small invited group .

6. **January 2022**: Phase 1 of the study was scheduled to begin on January 31, involving 50 severe and very severe ME patients plus 500 randomly selected participants from those who signed up online .

7. **2022**: Data analysis was expected to take several months, with initial results possibly available by mid-2023. Additional funding allowed for the inclusion of an extra 5,000 people who developed ME/CFS following COVID-19 .

For further details, you can visit the DecodeME page on MEpedia: https://me-pedia.org/wiki/DecodeME.
 
Back
Top Bottom