Back

Fine-Tuning the Llama2 Large Language Model Using Books on the Diagnosis and Treatment of Musculoskeletal System in Physical Therapy

Kim, J.-h.

2023-11-23 rehabilitation medicine and physical therapy
10.1101/2023.11.23.23298943 medRxiv
Show abstract

BackgroudGenerative language models (GLM) utilize machine learning algorithms to perform various tasks such as text generation, question response, and sentence completion by imitating the language that humans understand and use. PurposeThis study was to fine-tune the Llama2 language model using text data from books on the diagnosis and treatment of musculoskeletal system in physical therapy and compare it to the base model to determine its usability in medical fields. ResultsCompared to the base model, the fine-tuned model consistently generated answers specific to the musculoskeletal system diagnosis and treatment, demonstrating improved understanding of the specialized domain. ConclusionThe model fine-tuned for musculoskeletal diagnosis and treatment books provided more detailed information related to musculoskeletal topics, and the use of this fine-tuned model could be helpful in medical education and the acquisition of specialized knowledge.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

1
PLOS Digital Health
91 papers in training set
Top 0.1%
17.4%
2
Healthcare
16 papers in training set
Top 0.1%
10.0%
3
Frontiers in Digital Health
20 papers in training set
Top 0.1%
8.2%
4
F1000Research
79 papers in training set
Top 0.1%
7.1%
5
PLOS ONE
4510 papers in training set
Top 25%
6.8%
6
Scientific Reports
3102 papers in training set
Top 18%
6.3%
50% of probability mass above
7
BMC Medical Informatics and Decision Making
39 papers in training set
Top 0.9%
3.1%
8
International Journal of Medical Informatics
25 papers in training set
Top 0.5%
2.7%
9
Journal of NeuroEngineering and Rehabilitation
28 papers in training set
Top 0.5%
1.9%
10
Frontiers in Psychology
49 papers in training set
Top 0.5%
1.8%
11
Bioengineering
24 papers in training set
Top 0.5%
1.7%
12
BMC Medical Education
20 papers in training set
Top 0.5%
1.7%
13
Artificial Intelligence in Medicine
15 papers in training set
Top 0.3%
1.7%
14
iScience
1063 papers in training set
Top 18%
1.5%
15
JMIR Medical Informatics
17 papers in training set
Top 1.0%
1.3%
16
Journal of Clinical Medicine
91 papers in training set
Top 5%
1.2%
17
Frontiers in Public Health
140 papers in training set
Top 6%
1.2%
18
Data in Brief
13 papers in training set
Top 0.2%
1.1%
19
Computers in Biology and Medicine
120 papers in training set
Top 3%
1.1%
20
Journal of Medical Internet Research
85 papers in training set
Top 3%
1.1%
21
DIGITAL HEALTH
12 papers in training set
Top 0.5%
0.9%
22
Brain Sciences
52 papers in training set
Top 2%
0.8%
23
Annals of the New York Academy of Sciences
12 papers in training set
Top 0.1%
0.8%
24
BMJ Open
554 papers in training set
Top 13%
0.7%
25
MethodsX
14 papers in training set
Top 0.5%
0.7%
26
Heliyon
146 papers in training set
Top 7%
0.7%
27
Biology Methods and Protocols
53 papers in training set
Top 3%
0.7%
28
Annals of Biomedical Engineering
34 papers in training set
Top 1%
0.7%
29
BMC Bioinformatics
383 papers in training set
Top 8%
0.6%
30
PeerJ
261 papers in training set
Top 18%
0.6%