OncoSLM / README.md
amlopeza's picture
Update README.md
d22110f verified
|
Raw History Blame Contribute Delete
1.35 kB
metadata
license: cc-by-nc-4.0

Model Card for OncoSLM

What is OncoSLM

Model description

  • Developed by: Ana Maria Lopez @ZhangLab - Virginia Tech
  • Model type: Pretrained LLM on oncology RWE data and finetuned model on survival prediction in five different cancer types.
  • Pretrainind dataset: MSK-CHORD dataset - Targeted sequencing of 25040 tumors from 24950 patients via MSK-IMPACT, along with clinical annotations, some of which are derived from natural language processing (denoted NLP). For details on the dataset and licensing check MSK-CHORD
  • Repository:
  • Paper:
  • License: CC-BY-NC-ND-4.0

How to use?

from oncoslm import OncoSLM

device = 'cpu'
model = OncoSLM.from_pretrained('amlopeza/SurvLM')
model.to(device)

prompt = '''
Context: This is a cancer patient with the following characteristics:
 * Gender: Male
 * History of PDL-1: No
 * Age of the patient: 75
 * Cancer stage: Stage 4
 * Mutations history: PIK3C2G, FLT4, PIK3R3, BLM, MAP3K1, CCNE1, ATR, BRIP1, TSHR, SDHA, RET, PTPRD, FBXW7, NSD1, PDGFRA, AXIN2, TP53, SMARCA4, PTPRT, HGF, DOT1L, KDM5A, PTPRS, CREBBP, KMT2C, ATRX
 * HER2 status: No
 * Treatment history: CISPLATIN, ETOPOSIDE, CARBOPLATIN, PEMBROLIZUMAB
'''

model.predict(prompt)