TxAgent-Api / app.py
Ali2206's picture
Create app.py
f126604 verified
raw
history blame
16.9 kB
import os
import json
import shutil
import re
import gc
import time
from datetime import datetime
from typing import List, Tuple, Dict, Union, Optional
from fastapi import FastAPI, UploadFile, File, HTTPException
from fastapi.responses import FileResponse, JSONResponse
from fastapi.middleware.cors import CORSMiddleware
import pandas as pd
import pdfplumber
import torch
import matplotlib.pyplot as plt
from fpdf import FPDF
import unicodedata
import uvicorn
# === Configuration ===
persistent_dir = "/data/hf_cache"
model_cache_dir = os.path.join(persistent_dir, "txagent_models")
tool_cache_dir = os.path.join(persistent_dir, "tool_cache")
file_cache_dir = os.path.join(persistent_dir, "cache")
report_dir = os.path.join(persistent_dir, "reports")
for d in [model_cache_dir, tool_cache_dir, file_cache_dir, report_dir]:
os.makedirs(d, exist_ok=True)
os.environ["HF_HOME"] = model_cache_dir
os.environ["TRANSFORMERS_CACHE"] = model_cache_dir
current_dir = os.path.dirname(os.path.abspath(__file__))
src_path = os.path.abspath(os.path.join(current_dir, "src"))
sys.path.insert(0, src_path)
from txagent.txagent import TxAgent
MAX_MODEL_TOKENS = 131072
MAX_NEW_TOKENS = 4096
MAX_CHUNK_TOKENS = 8192
BATCH_SIZE = 1
PROMPT_OVERHEAD = 300
SAFE_SLEEP = 0.5
app = FastAPI(title="Clinical Patient Support System API",
description="API for analyzing and summarizing unstructured medical files",
version="1.0.0")
# CORS configuration
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
# Initialize agent at startup
agent = None
@app.on_event("startup")
async def startup_event():
global agent
agent = init_agent()
def estimate_tokens(text: str) -> int:
return len(text) // 4 + 1
def clean_response(text: str) -> str:
text = re.sub(r"\[.*?\]|\bNone\b", "", text, flags=re.DOTALL)
text = re.sub(r"\n{3,}", "\n\n", text)
return text.strip()
def remove_duplicate_paragraphs(text: str) -> str:
paragraphs = text.strip().split("\n\n")
seen = set()
unique_paragraphs = []
for p in paragraphs:
clean_p = p.strip()
if clean_p and clean_p not in seen:
unique_paragraphs.append(clean_p)
seen.add(clean_p)
return "\n\n".join(unique_paragraphs)
def extract_text_from_excel(path: str) -> str:
all_text = []
xls = pd.ExcelFile(path)
for sheet_name in xls.sheet_names:
try:
df = xls.parse(sheet_name).astype(str).fillna("")
except Exception:
continue
for _, row in df.iterrows():
non_empty = [cell.strip() for cell in row if cell.strip()]
if len(non_empty) >= 2:
text_line = " | ".join(non_empty)
if len(text_line) > 15:
all_text.append(f"[{sheet_name}] {text_line}")
return "\n".join(all_text)
def extract_text_from_csv(path: str) -> str:
all_text = []
try:
df = pd.read_csv(path).astype(str).fillna("")
except Exception:
return ""
for _, row in df.iterrows():
non_empty = [cell.strip() for cell in row if cell.strip()]
if len(non_empty) >= 2:
text_line = " | ".join(non_empty)
if len(text_line) > 15:
all_text.append(text_line)
return "\n".join(all_text)
def extract_text_from_pdf(path: str) -> str:
import logging
logging.getLogger("pdfminer").setLevel(logging.ERROR)
all_text = []
try:
with pdfplumber.open(path) as pdf:
for page in pdf.pages:
text = page.extract_text()
if text:
all_text.append(text.strip())
except Exception:
return ""
return "\n".join(all_text)
def extract_text(file_path: str) -> str:
if file_path.endswith(".xlsx"):
return extract_text_from_excel(file_path)
elif file_path.endswith(".csv"):
return extract_text_from_csv(file_path)
elif file_path.endswith(".pdf"):
return extract_text_from_pdf(file_path)
else:
return ""
def split_text(text: str, max_tokens=MAX_CHUNK_TOKENS) -> List[str]:
effective_limit = max_tokens - PROMPT_OVERHEAD
chunks, current, current_tokens = [], [], 0
for line in text.split("\n"):
tokens = estimate_tokens(line)
if current_tokens + tokens > effective_limit:
if current:
chunks.append("\n".join(current))
current, current_tokens = [line], tokens
else:
current.append(line)
current_tokens += tokens
if current:
chunks.append("\n".join(current))
return chunks
def batch_chunks(chunks: List[str], batch_size: int = BATCH_SIZE) -> List[List[str]]:
return [chunks[i:i+batch_size] for i in range(0, len(chunks), batch_size)]
def build_prompt(chunk: str) -> str:
return f"""### Unstructured Clinical Records\n\nAnalyze the clinical notes below and summarize with:\n- Diagnostic Patterns\n- Medication Issues\n- Missed Opportunities\n- Inconsistencies\n- Follow-up Recommendations\n\n---\n\n{chunk}\n\n---\nRespond concisely in bullet points with clinical reasoning."""
def init_agent() -> TxAgent:
tool_path = os.path.join(tool_cache_dir, "new_tool.json")
if not os.path.exists(tool_path):
shutil.copy(os.path.abspath("data/new_tool.json"), tool_path)
agent = TxAgent(
model_name="mims-harvard/TxAgent-T1-Llama-3.1-8B",
rag_model_name="mims-harvard/ToolRAG-T1-GTE-Qwen2-1.5B",
tool_files_dict={"new_tool": tool_path},
force_finish=True,
enable_checker=True,
step_rag_num=4,
seed=100
)
agent.init_model()
return agent
def analyze_batches(agent, batches: List[List[str]]) -> List[str]:
results = []
for batch in batches:
prompt = "\n\n".join(build_prompt(chunk) for chunk in batch)
try:
batch_response = ""
for r in agent.run_gradio_chat(
message=prompt,
history=[],
temperature=0.0,
max_new_tokens=MAX_NEW_TOKENS,
max_token=MAX_MODEL_TOKENS,
call_agent=False,
conversation=[]
):
if isinstance(r, str):
batch_response += r
elif isinstance(r, list):
for m in r:
if hasattr(m, "content"):
batch_response += m.content
elif hasattr(r, "content"):
batch_response += r.content
results.append(clean_response(batch_response))
time.sleep(SAFE_SLEEP)
except Exception as e:
results.append(f"❌ Batch failed: {str(e)}")
time.sleep(SAFE_SLEEP * 2)
torch.cuda.empty_cache()
gc.collect()
return results
def generate_final_summary(agent, combined: str) -> str:
combined = remove_duplicate_paragraphs(combined)
final_prompt = f"""
You are an expert clinical summarizer. Analyze the following summaries carefully and generate a **single final concise structured medical report**, avoiding any repetition or redundancy.
Summaries:
{combined}
Respond with:
- Diagnostic Patterns
- Medication Issues
- Missed Opportunities
- Inconsistencies
- Follow-up Recommendations
Avoid repeating the same points multiple times.
""".strip()
final_response = ""
for r in agent.run_gradio_chat(
message=final_prompt,
history=[],
temperature=0.0,
max_new_tokens=MAX_NEW_TOKENS,
max_token=MAX_MODEL_TOKENS,
call_agent=False,
conversation=[]
):
if isinstance(r, str):
final_response += r
elif isinstance(r, list):
for m in r:
if hasattr(m, "content"):
final_response += m.content
elif hasattr(r, "content"):
final_response += r.content
final_response = clean_response(final_response)
final_response = remove_duplicate_paragraphs(final_response)
return final_response
def remove_non_ascii(text):
return ''.join(c for c in text if ord(c) < 256)
def generate_pdf_report_with_charts(summary: str, report_path: str, detailed_batches: List[str] = None):
chart_dir = os.path.join(os.path.dirname(report_path), "charts")
os.makedirs(chart_dir, exist_ok=True)
# Prepare static data
categories = ['Diagnostics', 'Medications', 'Missed', 'Inconsistencies', 'Follow-up']
values = [4, 2, 3, 1, 5]
# === Static Charts ===
chart_paths = []
def save_chart(fig_func, filename):
path = os.path.join(chart_dir, filename)
fig_func()
plt.tight_layout()
plt.savefig(path)
plt.close()
chart_paths.append((filename.split('.')[0].replace('_', ' ').title(), path))
save_chart(lambda: plt.bar(categories, values), "bar_chart.png")
save_chart(lambda: plt.pie(values, labels=categories, autopct='%1.1f%%'), "pie_chart.png")
save_chart(lambda: plt.plot(categories, values, marker='o'), "trend_chart.png")
save_chart(lambda: plt.barh(categories, values), "horizontal_bar_chart.png")
# Radar chart
import numpy as np
labels = np.array(categories)
stats = np.array(values)
angles = np.linspace(0, 2 * np.pi, len(labels), endpoint=False).tolist()
stats = np.concatenate((stats, [stats[0]]))
angles += angles[:1]
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(polar=True))
ax.plot(angles, stats, marker='o')
ax.fill(angles, stats, alpha=0.25)
ax.set_yticklabels([])
ax.set_xticks(angles[:-1])
ax.set_xticklabels(labels)
ax.set_title('Radar Chart: Clinical Focus')
radar_path = os.path.join(chart_dir, "radar_chart.png")
plt.tight_layout()
plt.savefig(radar_path)
plt.close()
chart_paths.append(("Radar Chart: Clinical Focus", radar_path))
# === Dynamic Chart: Drug Frequency ===
drug_counter = {}
if detailed_batches:
for batch in detailed_batches:
lines = batch.split("\n")
for line in lines:
match = re.search(r"(?i)medication[s]?:\s*(.+)", line)
if match:
items = re.split(r"[,;]", match.group(1))
for item in items:
drug = item.strip().title()
if len(drug) > 2:
drug_counter[drug] = drug_counter.get(drug, 0) + 1
if drug_counter:
drugs, freqs = zip(*sorted(drug_counter.items(), key=lambda x: x[1], reverse=True)[:10])
plt.figure(figsize=(6, 4))
plt.bar(drugs, freqs)
plt.xticks(rotation=45, ha='right')
plt.title('Top Medications Frequency')
drug_chart_path = os.path.join(chart_dir, "drug_frequency_chart.png")
plt.tight_layout()
plt.savefig(drug_chart_path)
plt.close()
chart_paths.append(("Top Medications Frequency", drug_chart_path))
# === PDF ===
pdf_path = report_path.replace('.md', '.pdf')
pdf = FPDF()
pdf.set_auto_page_break(auto=True, margin=20)
def add_section_title(pdf, title):
pdf.set_fill_color(230, 230, 230)
pdf.set_font("Arial", 'B', 14)
pdf.cell(0, 10, remove_non_ascii(title), ln=True, fill=True)
pdf.ln(3)
def add_footer(pdf):
pdf.set_y(-15)
pdf.set_font('Arial', 'I', 8)
pdf.set_text_color(150, 150, 150)
pdf.cell(0, 10, f"Page {pdf.page_no()}", align='C')
# Title Page
pdf.add_page()
pdf.set_font("Arial", 'B', 26)
pdf.set_text_color(0, 70, 140)
pdf.cell(0, 20, remove_non_ascii("Final Medical Report"), ln=True, align='C')
pdf.set_text_color(0, 0, 0)
pdf.set_font("Arial", '', 13)
pdf.cell(0, 10, datetime.now().strftime("Generated on %B %d, %Y at %H:%M"), ln=True, align='C')
pdf.ln(15)
pdf.set_font("Arial", '', 11)
pdf.set_fill_color(245, 245, 245)
pdf.multi_cell(0, 9, remove_non_ascii(
"This report contains a professional summary of clinical observations, potential inconsistencies, and follow-up recommendations based on the uploaded medical document."
), border=1, fill=True, align="J")
add_footer(pdf)
# Final Summary
pdf.add_page()
add_section_title(pdf, "Final Summary")
pdf.set_font("Arial", '', 11)
for line in summary.split("\n"):
clean_line = remove_non_ascii(line.strip())
if clean_line:
pdf.multi_cell(0, 8, txt=clean_line)
add_footer(pdf)
# Charts Section
pdf.add_page()
add_section_title(pdf, "Statistical Overview")
for title, path in chart_paths:
pdf.set_font("Arial", 'B', 12)
pdf.cell(0, 9, remove_non_ascii(title), ln=True)
pdf.image(path, w=170)
pdf.ln(6)
add_footer(pdf)
# Detailed Tool Outputs
if detailed_batches:
pdf.add_page()
add_section_title(pdf, "Detailed Tool Insights")
for idx, detail in enumerate(detailed_batches):
pdf.set_font("Arial", 'B', 12)
pdf.cell(0, 9, remove_non_ascii(f"Tool Output #{idx + 1}"), ln=True)
pdf.set_font("Arial", '', 11)
for line in remove_non_ascii(detail).split("\n"):
pdf.multi_cell(0, 8, txt=line.strip())
pdf.ln(3)
add_footer(pdf)
pdf.output(pdf_path)
return pdf_path
@app.post("/analyze", summary="Analyze medical document", response_description="Returns analysis results")
async def analyze_document(file: UploadFile = File(...)):
"""
Analyze a medical document (PDF, Excel, or CSV) and return a structured analysis.
Args:
file: The medical document to analyze (PDF, Excel, or CSV format)
Returns:
JSONResponse: Contains analysis results and report download path
"""
start_time = time.time()
try:
# Save the uploaded file temporarily
temp_path = os.path.join(file_cache_dir, file.filename)
with open(temp_path, "wb") as f:
f.write(await file.read())
extracted = extract_text(temp_path)
if not extracted:
raise HTTPException(status_code=400, detail="Could not extract text from the file")
chunks = split_text(extracted)
batches = batch_chunks(chunks, batch_size=BATCH_SIZE)
batch_results = analyze_batches(agent, batches)
all_tool_outputs = batch_results.copy()
valid = [res for res in batch_results if not res.startswith("❌")]
if not valid:
raise HTTPException(status_code=400, detail="No valid analysis results were generated")
summary = generate_final_summary(agent, "\n\n".join(valid))
# Generate report files
report_filename = f"report_{datetime.now().strftime('%Y%m%d_%H%M%S')}"
report_path = os.path.join(report_dir, f"{report_filename}.md")
with open(report_path, 'w', encoding='utf-8') as f:
f.write(f"# Final Medical Report\n\n{summary}")
pdf_path = generate_pdf_report_with_charts(summary, report_path, detailed_batches=all_tool_outputs)
end_time = time.time()
elapsed_time = end_time - start_time
# Clean up temp file
os.remove(temp_path)
return JSONResponse({
"status": "success",
"summary": summary,
"report_path": f"/reports/{os.path.basename(pdf_path)}",
"processing_time": f"{elapsed_time:.2f} seconds",
"detailed_outputs": all_tool_outputs
})
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
@app.get("/reports/{filename}", response_class=FileResponse)
async def download_report(filename: str):
"""
Download a generated report PDF file.
Args:
filename: The name of the report file to download
Returns:
FileResponse: The PDF file for download
"""
file_path = os.path.join(report_dir, filename)
if not os.path.exists(file_path):
raise HTTPException(status_code=404, detail="Report not found")
return FileResponse(file_path, media_type='application/pdf', filename=filename)
@app.get("/status")
async def service_status():
"""
Check the service status and version information.
Returns:
JSONResponse: Service status information
"""
return JSONResponse({
"status": "running",
"version": "1.0.0",
"model": "mims-harvard/TxAgent-T1-Llama-3.1-8B",
"rag_model": "mims-harvard/ToolRAG-T1-GTE-Qwen2-1.5B",
"max_tokens": MAX_MODEL_TOKENS,
"supported_file_types": [".pdf", ".xlsx", ".csv"]
})
if __name__ == "__main__":
uvicorn.run(app, host="0.0.0.0", port=7860)