ChatGPT fails challenging the recent ESCMID brain abscess guideline
© 2024. The Author(s)..
BACKGROUND: With artificial intelligence (AI) on the rise, it remains unclear if AI is able to professionally evaluate medical research and give scientifically valid recommendations.
AIM: This study aimed to assess the accuracy of ChatGPT's responses to ten key questions on brain abscess diagnostics and treatment in comparison to the guideline recently published by the European Society for Clinical Microbiology and Infectious Diseases (ESCMID).
METHODS: All ten PECO (Population, Exposure, Comparator, Outcome) questions which had been developed during the guideline process were presented directly to ChatGPT. Next, ChatGPT was additionally fed with data from studies selected for each PECO question by the ESCMID committee. AI's responses were subsequently compared with the recommendations of the ESCMID guideline.
RESULTS: For 17 out of 20 challenges, ChatGPT was able to give recommendations on the management of patients with brain abscess, including grade of evidence and strength of recommendation. Without data prompting, 70% of questions were answered very similar to the guideline recommendation. In the answers that differed from the guideline recommendations, no patient hazard was present. Data input slightly improved the clarity of ChatGPT's recommendations, but, however, led to less correct answers including two recommendations that directly contradicted the guideline, being associated with the possibility of a hazard to the patient.
CONCLUSION: ChatGPT seems to be able to rapidly gather information on brain abscesses and give recommendations on key questions about their management in most cases. Nevertheless, single responses could possibly harm the patients. Thus, the expertise of an expert committee remains inevitable.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2024 |
|---|---|
| Erschienen: |
2024 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:271 |
|---|---|
| Enthalten in: |
Journal of neurology - 271(2024), 4 vom: 17. März, Seite 2086-2101 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Dyckhoff-Shen, Susanne [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
|---|
| Anmerkungen: |
Date Completed 28.03.2024 Date Revised 30.03.2024 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1007/s00415-023-12168-1 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM367704102 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM367704102 | ||
| 003 | DE-627 | ||
| 005 | 20240331000907.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 240128s2024 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s00415-023-12168-1 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1357.xml |
| 035 | |a (DE-627)NLM367704102 | ||
| 035 | |a (NLM)38279999 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Dyckhoff-Shen, Susanne |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a ChatGPT fails challenging the recent ESCMID brain abscess guideline |
| 264 | 1 | |c 2024 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 28.03.2024 | ||
| 500 | |a Date Revised 30.03.2024 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © 2024. The Author(s). | ||
| 520 | |a BACKGROUND: With artificial intelligence (AI) on the rise, it remains unclear if AI is able to professionally evaluate medical research and give scientifically valid recommendations | ||
| 520 | |a AIM: This study aimed to assess the accuracy of ChatGPT's responses to ten key questions on brain abscess diagnostics and treatment in comparison to the guideline recently published by the European Society for Clinical Microbiology and Infectious Diseases (ESCMID) | ||
| 520 | |a METHODS: All ten PECO (Population, Exposure, Comparator, Outcome) questions which had been developed during the guideline process were presented directly to ChatGPT. Next, ChatGPT was additionally fed with data from studies selected for each PECO question by the ESCMID committee. AI's responses were subsequently compared with the recommendations of the ESCMID guideline | ||
| 520 | |a RESULTS: For 17 out of 20 challenges, ChatGPT was able to give recommendations on the management of patients with brain abscess, including grade of evidence and strength of recommendation. Without data prompting, 70% of questions were answered very similar to the guideline recommendation. In the answers that differed from the guideline recommendations, no patient hazard was present. Data input slightly improved the clarity of ChatGPT's recommendations, but, however, led to less correct answers including two recommendations that directly contradicted the guideline, being associated with the possibility of a hazard to the patient | ||
| 520 | |a CONCLUSION: ChatGPT seems to be able to rapidly gather information on brain abscesses and give recommendations on key questions about their management in most cases. Nevertheless, single responses could possibly harm the patients. Thus, the expertise of an expert committee remains inevitable | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a AI | |
| 650 | 4 | |a Brain abscess | |
| 650 | 4 | |a ChatGPT | |
| 650 | 4 | |a Guideline | |
| 700 | 1 | |a Koedel, Uwe |e verfasserin |4 aut | |
| 700 | 1 | |a Brouwer, Matthijs C |e verfasserin |4 aut | |
| 700 | 1 | |a Bodilsen, Jacob |e verfasserin |4 aut | |
| 700 | 1 | |a Klein, Matthias |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Journal of neurology |d 1974 |g 271(2024), 4 vom: 17. März, Seite 2086-2101 |w (DE-627)NLM000508543 |x 1432-1459 |7 nnns |
| 773 | 1 | 8 | |g volume:271 |g year:2024 |g number:4 |g day:17 |g month:03 |g pages:2086-2101 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1007/s00415-023-12168-1 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 271 |j 2024 |e 4 |b 17 |c 03 |h 2086-2101 | ||