bolade/bio-performx
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: Enhance context generation and report generation services with improved data handling and structure

Browse Source
This commit is contained in:
bolade
2025-10-04 10:25:10 +01:00
parent d66f3fd18b
commit 358898b7db
7 changed files with 163 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/services/__pycache__/context_generator.cpython-312.pyc
BIN
View File
Binary file not shown.
app/services/__pycache__/graph_generator.cpython-312.pyc
BIN
View File
Binary file not shown.
app/services/__pycache__/report_generator.cpython-312.pyc
BIN
View File
Binary file not shown.
app/services/context_generator.py
+142 -42
View File
@@ -6,7 +6,6 @@ of the medical report. It performs analysis on Pnoe, Spirometry, and SECA data.
"""
from datetime import datetime
from pathlib import Path
from typing import Dict, List, Optional, Tuple
import pandas as pd
@@ -35,28 +34,59 @@ class ContextGenerator:
def _preprocess_pnoe_data(self):
"""Apply preprocessing steps to Pnoe data"""
self.pnoe_df = self.pnoe_df.apply(pd.to_numeric, errors="ignore")
self.pnoe_df["VO2 Pulse"] = self.pnoe_df["VO2(ml/min)"] / self.pnoe_df["HR(bpm)"]
self.pnoe_df["VO2 Breath"] = self.pnoe_df["VO2(ml/min)"] / self.pnoe_df["BF(bpm)"]
self.pnoe_df["CHO"] = self.pnoe_df["EE(kcal/min)"] * self.pnoe_df["CARBS(%)"] / 100
self.pnoe_df["FAT"] = self.pnoe_df["EE(kcal/min)"] * self.pnoe_df["FAT(%)"] / 100
# Convert numeric columns
for col in self.pnoe_df.columns:
try:
self.pnoe_df[col] = pd.to_numeric(self.pnoe_df[col])
except (ValueError, TypeError):
pass
self.pnoe_df["VO2 Pulse"] = (
self.pnoe_df["VO2(ml/min)"] / self.pnoe_df["HR(bpm)"]
)
self.pnoe_df["VO2 Breath"] = (
self.pnoe_df["VO2(ml/min)"] / self.pnoe_df["BF(bpm)"]
)
self.pnoe_df["CHO"] = (
self.pnoe_df["EE(kcal/min)"] * self.pnoe_df["CARBS(%)"] / 100
)
self.pnoe_df["FAT"] = (
self.pnoe_df["EE(kcal/min)"] * self.pnoe_df["FAT(%)"] / 100
)
window_size = 10
columns_to_smooth = ["VO2(ml/min)", "VCO2(ml/min)", "HR(bpm)", "VT(l)", "BF(bpm)", "VE(l/min)", "VO2 Pulse", "VO2 Breath", "CHO", "FAT"]
columns_to_smooth = [
"VO2(ml/min)",
"VCO2(ml/min)",
"HR(bpm)",
"VT(l)",
"BF(bpm)",
"VE(l/min)",
"VO2 Pulse",
"VO2 Breath",
"CHO",
"FAT",
]
for col in columns_to_smooth:
if col in self.pnoe_df.columns:
self.pnoe_df[f"{col}_smoothed"] = self.pnoe_df[col].rolling(window=window_size, min_periods=1).mean()
self.pnoe_df[f"{col}_smoothed"] = (
self.pnoe_df[col].rolling(window=window_size, min_periods=1).mean()
)
def extract_patient_info(self, patient_name: str) -> Dict:
"""Extract patient information from SECA dataset"""
if self.seca_df is not None:
patient_data = self.seca_df[self.seca_df["LastName"].str.contains(patient_name, case=False, na=False)]
patient_data = self.seca_df[
self.seca_df["LastName"].str.contains(
patient_name, case=False, na=False
)
]
if not patient_data.empty:
row = patient_data.iloc[0]
weight_kg = float(row.get("Weight", 0))
fat_pct = float(row.get("Adult_FMP", 0))
self.patient_info = {
"name": f"{row.get('FirstName', '')} {row.get('LastName', '')}",
"first_name": row.get("FirstName", ""),
@@ -75,9 +105,11 @@ class ContextGenerator:
"""Calculate spirometry-related metrics"""
metrics = {}
for param in ["FVC", "FEV1", "FEV1/FVC%"]:
row = self.spirometry_df.loc[self.spirometry_df["Parameters"].str.strip() == param]
row = self.spirometry_df.loc[
self.spirometry_df["Parameters"].str.strip() == param
]
if not row.empty:
param_key = param.lower().replace('/', '_').replace('%', '_pct')
param_key = param.lower().replace("/", "_").replace("%", "_pct")
metrics[f"{param_key}_best"] = row["Best"].values[0]
metrics[f"{param_key}_pred"] = row["%Pred."].values[0]
return metrics
@@ -87,21 +119,21 @@ class ContextGenerator:
metrics = {}
metrics["vo2_max"] = self.pnoe_df["VO2(ml/min)_smoothed"].max()
metrics["vo2_max_per_kg"] = metrics["vo2_max"] / self.patient_info["weight"]
peak_vt_idx = self.pnoe_df["VT(l)_smoothed"].idxmax()
peak_vt_row = self.pnoe_df.loc[peak_vt_idx]
metrics["peak_vt"] = peak_vt_row["VT(l)_smoothed"]
metrics["peak_vt_hr"] = peak_vt_row["HR(bpm)_smoothed"]
fat_max_idx = self.pnoe_df["FAT_smoothed"].idxmax()
fat_max_row = self.pnoe_df.loc[fat_max_idx]
metrics["fat_max_value"] = fat_max_row["FAT_smoothed"]
metrics["fat_max_hr"] = fat_max_row["HR(bpm)_smoothed"]
vt1, vt2 = self._detect_thresholds()
metrics["vt1"] = vt1
metrics["vt2"] = vt2
zones = self._calculate_hr_zones(vt1, vt2, fat_max_row)
metrics.update(zones)
return metrics
@@ -110,25 +142,35 @@ class ContextGenerator:
"""Detect VT1 and VT2 thresholds"""
condition = self.pnoe_df["CHO_smoothed"] > self.pnoe_df["FAT_smoothed"]
crossover_indices = condition[condition].index
vt1 = None
if len(crossover_indices) > 0:
vt1_idx = crossover_indices[0]
vt1_row = self.pnoe_df.loc[vt1_idx]
vt1 = {"HeartRate": vt1_row["HR(bpm)_smoothed"], "Speed": vt1_row["Speed"], "Time": vt1_row["T(sec)"]}
vt1 = {
"HeartRate": vt1_row["HR(bpm)_smoothed"],
"Speed": vt1_row["Speed"],
"Time": vt1_row["T(sec)"],
}
ve_slope = self.pnoe_df["VE(l/min)_smoothed"].diff()
second_derivative = ve_slope.diff()
vt2_idx = second_derivative.idxmax()
vt2 = None
if pd.notna(vt2_idx):
vt2_row = self.pnoe_df.loc[vt2_idx]
vt2 = {"HeartRate": vt2_row["HR(bpm)_smoothed"], "Speed": vt2_row["Speed"], "Time": vt2_row["T(sec)"]}
vt2 = {
"HeartRate": vt2_row["HR(bpm)_smoothed"],
"Speed": vt2_row["Speed"],
"Time": vt2_row["T(sec)"],
}
return vt1, vt2
def _calculate_hr_zones(self, vt1: Optional[Dict], vt2: Optional[Dict], fat_max_row: pd.Series) -> Dict:
def _calculate_hr_zones(
self, vt1: Optional[Dict], vt2: Optional[Dict], fat_max_row: pd.Series
) -> Dict:
"""Calculate heart rate zones based on thresholds"""
zones = {}
if vt1 and vt2:
@@ -137,7 +179,7 @@ class ContextGenerator:
zone_3_start = vt1["HeartRate"]
zone_4_start = vt2["HeartRate"] - 10
zone_5_start = vt2["HeartRate"] + 10
zones["zone1_bpm"] = f"{int(zone_1_start)}-{int(zone_2_start)}bpm"
zones["zone2_bpm"] = f"{int(zone_2_start)}-{int(vt1['HeartRate'])}bpm"
zones["zone3_bpm"] = f"{int(zone_3_start)}-{int(zone_4_start)}bpm"
@@ -152,29 +194,87 @@ class ContextGenerator:
zones["zone5_bpm"] = f"{int(max_hr * 0.95)}+bpm"
return zones
def generate_all_contexts(self, patient_name: str, graphs: Dict[str, str]) -> List[Dict]:
def generate_all_contexts(
self, patient_name: str, graphs: Dict[str, str]
) -> List[Dict]:
"""Main method to generate all page contexts"""
self.extract_patient_info(patient_name)
spirometry_metrics = self.calculate_spirometry_metrics()
pnoe_metrics = self.calculate_pnoe_metrics()
contexts = []
contexts.append({"name": self.patient_info["name"], "surname": self.patient_info["last_name"], "date": datetime.now().strftime("%B %d, %Y")})
contexts.append({"patient_name": self.patient_info["name"], "test_date": datetime.now().strftime("%B %d, %Y")})
contexts.append(
{
"name": self.patient_info["name"],
"surname": self.patient_info["last_name"],
"date": datetime.now().strftime("%B %d, %Y"),
}
)
contexts.append(
{
"patient_name": self.patient_info["name"],
"test_date": datetime.now().strftime("%B %d, %Y"),
}
)
for i in range(4):
contexts.append({"patient_name": self.patient_info["name"], "page_number": i + 3})
contexts.append(
{"patient_name": self.patient_info["name"], "page_number": i + 3}
)
fev1_percentage = 0
if spirometry_metrics.get("fvc_best"):
fev1_percentage = (pnoe_metrics["peak_vt"] / spirometry_metrics["fvc_best"]) * 100
contexts.append({"peak_vt": f"{pnoe_metrics['peak_vt']:.2f}", "peak_vt_bpm": f"{int(pnoe_metrics['peak_vt_hr'])}", "fev1_percentage": f"{fev1_percentage:.1f}", "lung_analysis_chart": graphs.get("spirometry_chart", ""), "respiratory_analysis_chart": graphs.get("respiratory", "")})
contexts.append({"vo2_max_value": f"{pnoe_metrics['vo2_max_per_kg']:.1f}", "age_range": f"{self.patient_info['age'] // 10 * 10}-{self.patient_info['age'] // 10 * 10 + 9}", "zone1_bpm": pnoe_metrics.get("zone1_bpm", ""), "zone2_bpm": pnoe_metrics.get("zone2_bpm", ""), "zone3_bpm": pnoe_metrics.get("zone3_bpm", ""), "zone4_bpm": pnoe_metrics.get("zone4_bpm", ""), "zone5_bpm": pnoe_metrics.get("zone5_bpm", ""), "vo2_pulse_chart": graphs.get("vo2_pulse", "")})
contexts.append({"fat_max_value": f"{pnoe_metrics['fat_max_value']:.2f}", "fat_max_hr": f"{int(pnoe_metrics['fat_max_hr'])}", "fuel_utilization_chart": graphs.get("fuel_utilization", ""), "fat_metabolism_chart": graphs.get("fat_metabolism", "")})
contexts.append({"fat_percentage": f"{self.patient_info['fat_percentage']:.1f}", "fat_mass_lbs": f"{self.patient_info['fat_mass_lbs']:.1f}", "lean_mass_lbs": f"{self.patient_info['lean_mass_lbs']:.1f}", "body_composition_chart": graphs.get("body_composition", ""), "body_fat_percent_chart": graphs.get("body_fat_percent", "")})
fev1_percentage = (
pnoe_metrics["peak_vt"] / spirometry_metrics["fvc_best"]
) * 100
contexts.append(
{
"peak_vt": f"{pnoe_metrics['peak_vt']:.2f}",
"peak_vt_bpm": f"{int(pnoe_metrics['peak_vt_hr'])}",
"fev1_percentage": f"{fev1_percentage:.1f}",
"lung_analysis_chart": graphs.get("spirometry_chart", ""),
"respiratory_analysis_chart": graphs.get("respiratory", ""),
}
)
contexts.append(
{
"vo2_max_value": f"{pnoe_metrics['vo2_max_per_kg']:.1f}",
"age_range": f"{self.patient_info['age'] // 10 * 10}-{self.patient_info['age'] // 10 * 10 + 9}",
"zone1_bpm": pnoe_metrics.get("zone1_bpm", ""),
"zone2_bpm": pnoe_metrics.get("zone2_bpm", ""),
"zone3_bpm": pnoe_metrics.get("zone3_bpm", ""),
"zone4_bpm": pnoe_metrics.get("zone4_bpm", ""),
"zone5_bpm": pnoe_metrics.get("zone5_bpm", ""),
"vo2_pulse_chart": graphs.get("vo2_pulse", ""),
}
)
contexts.append(
{
"fat_max_value": f"{pnoe_metrics['fat_max_value']:.2f}",
"fat_max_hr": f"{int(pnoe_metrics['fat_max_hr'])}",
"fuel_utilization_chart": graphs.get("fuel_utilization", ""),
"fat_metabolism_chart": graphs.get("fat_metabolism", ""),
}
)
contexts.append(
{
"fat_percentage": f"{self.patient_info['fat_percentage']:.1f}",
"fat_mass_lbs": f"{self.patient_info['fat_mass_lbs']:.1f}",
"lean_mass_lbs": f"{self.patient_info['lean_mass_lbs']:.1f}",
"body_composition_chart": graphs.get("body_composition", ""),
"body_fat_percent_chart": graphs.get("body_fat_percent", ""),
}
)
for i in range(9):
contexts.append({"patient_name": self.patient_info["name"], "page_number": i + 11, "vo2_breath_chart": graphs.get("vo2_breath", ""), "recovery_chart": graphs.get("recovery", "")})
contexts.append(
{
"patient_name": self.patient_info["name"],
"page_number": i + 11,
"vo2_breath_chart": graphs.get("vo2_breath", ""),
"recovery_chart": graphs.get("recovery", ""),
}
)
return contexts
app/services/graph_generator.py
+3
View File
@@ -8,6 +8,9 @@ Based on the analysis notebooks in services_dfdf/.
import base64
from pathlib import Path
import matplotlib
matplotlib.use("Agg") # Use non-interactive backend
import matplotlib.pyplot as plt
import matplotlib.transforms as mtransforms
import numpy as np
app/services/report_generator.py
+11 -5
View File
@@ -11,10 +11,9 @@ from typing import Any, Dict, List
import pandas as pd
from jinja2 import Environment, FileSystemLoader
from playwright.sync_api import sync_playwright
from app.services.context_generator import ContextGenerator
from app.services.graph_generator import GraphGenerator
from app.services.spirometry_table_extractor import extract_spirometry_table_from_pdf
from services.context_generator import ContextGenerator
from services.graph_generator import GraphGenerator
from services.spirometry_table_extractor import extract_spirometry_table_from_pdf
class ReportGeneratorService:
@@ -61,7 +60,13 @@ class ReportGeneratorService:
"""
# Load data
df = pd.read_csv(pnoe_csv_path, delimiter=";")
df = df.apply(pd.to_numeric, errors="ignore")
# Convert numeric columns (updated approach)
for col in df.columns:
try:
df[col] = pd.to_numeric(df[col])
except (ValueError, TypeError):
pass # Keep as-is if not numeric
# Calculate derived columns
df["VO2 Pulse"] = df["VO2(ml/min)"] / df["HR(bpm)"]
@@ -395,6 +400,7 @@ class ReportGeneratorService:
)
report_path = self.reports_dir / output_filename
print(f"Generating PDF report at {report_path}")
self.html_to_pdf(html_content, str(report_path))
return {