Hot off the presses is an open source implementation of the Framingham 10 Year Heart Attack Risk Calculator. This software package consists of a command-line utility and a Python library for assessing the risk of a heart attack in individuals over the next 10 years. The variables used are sex, age, total cholesterol, HDL cholesterol, hypertension medication status, smoking status, and systolic blood pressure. For more information about the Framingham Heart Study click here. This software released under the Apache2 open source license.
Thanks to Ted Eytan (on Twitter @tedeytan) for pointing me to this in the first place.
You can get the full source code distribution and detailed usage instructions on Videntity’s GitHub Page.
You can install this package directly through pip by typing the following on your command line.
> sudo pip install python-framingham10yr
Now to use the function, you can do something like this.
>>> from framingham10yr.framingham10yr import framingham_10year_risk
>>> framingham_10year_risk("male", 36, 140, 80, 120, "no", "no")
{'status': 200, 'total_cholesterol': 140, 'smoker': False, 'blood_pressure_med_treatment': False, 'message': 'OK', 'sex': 'male', 'percent_risk': '<1%', 'age': 36, 'hdl_cholesterol': 80, 'systolic_blood_pressure': 120, 'points': -5}
>>>
For your inspection and edification, below is implementation of the algorithm itself.
!/usr/bin/env python
# -*- coding: utf-8 -*-
# vim: ai ts=4 sts=4 et sw=
# Copyright Videnty Systems Inc, 2012 videntity.com
# Apache 2 License
import sys, json
GENDER_CHOICES = (('MALE', 'MALE'),
('FEMALE', 'FEMALE'),)
SMOKER_CHOICES = ((True, 'Yes'),
(False, 'No'),)
def framingham_10year_risk(sex, age, total_cholesterol, hdl_cholesterol,
systolic_blood_pressure, smoker,
blood_pressure_med_treatment):
"""Requires:
sex - "male" or "female" string
age - string or int
total_cholesterol - sting or int
hdl_cholesterol - int
systolic_blood_pressure - int
smoker - True or False. Also accepts 1 or 0 as
a string or an int
blood_pressure_med_treatment - True or False. Also accepts 1 or 0
as a string or an int
"""
#be liberal in what we accept...massage the input
if sex in ("MALE", "m", "M", "boy", "xy", "male", "Male"):
sex = "male"
if sex in ("FEMALE", "f", "F", "girl", "xx", "female", "Female"):
sex = "female"
if smoker in ("yes", "Y", "y", "YES", "true", "t", "True", True, 1, "1"):
smoker=True
if smoker in ("no", "NO", "N", "n", "false", "f", "False", False, 0, "0"):
smoker=False
if blood_pressure_med_treatment in ("yes", "Y", "y", "YES", "true", "t",
"True", True, 1, "1"):
blood_pressure_med_treatment = True
if blood_pressure_med_treatment in ("no", "NO", "N", "n", "false", "f",
"False", False, 0, "0"):
blood_pressure_med_treatment = False
#intialize some things -----------------------------------------------------
errors = [] #a list of errors
points = 0
age = int(age)
total_cholesterol = int(total_cholesterol)
hdl_cholesterol = int(hdl_cholesterol)
systolic_blood_pressure = int(systolic_blood_pressure)
try:
blood_pressure_med_treatment = bool(int(blood_pressure_med_treatment))
except(ValueError):
errors.append("Blood pressure medication treatment must be set to True, False, 1, 0, yes, or no.")
try:
smoker = bool(int(smoker))
except(ValueError):
errors.append("Smoker must be set to True, False, 1, 0, yes, or no.")
# Intitalize our response dictionary
response = {"status": 200,
"sex":sex,
"message": "OK",
"age": age,
"total_cholesterol": total_cholesterol,
"hdl_cholesterol" : hdl_cholesterol,
"systolic_blood_pressure": systolic_blood_pressure,
"smoker": smoker,
"blood_pressure_med_treatment": blood_pressure_med_treatment,
}
#run some sanity checks ----------------------------------------------------
if not 20 <= age <=79:
errors.append("Age must be within the range of 20 to 79.")
if not 130 <= total_cholesterol <= 320:
errors.append("Total cholesterol must be within the range of 130 to 320.")
if not 20 <= hdl_cholesterol <= 100:
errors.append("HDL cholesterol must be within the range of 20 to 100.")
if not 90 <= systolic_blood_pressure<= 200:
errors.append("Systolic blood pressure must be within the range of 90 to 200.")
if sex.lower() not in ('male', 'female'):
errors.append("Sex must be male or female.")
#Process males -----------------------------------------------------------
if sex.lower()=="male":
# Age - male
if 20 <= age <= 34:
points-=9
if 35 <= age <= 39:
points-=4
if 40 <= age <= 44:
points-=0
if 45 <= age <= 49:
points+=3
if 50 <= age <= 54:
points+=6
if 55 <= age <= 59:
points+=8
if 60 <= age <= 64:
points+=10
if 65 <= age <= 69:
points+=12
if 70 <= age <= 74:
points+=14
if 75 <= age <= 79:
points+=16
#Total cholesterol, mg/dL - Male ------------------------
if 20 <= age <= 39:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=4
if 200 <= total_cholesterol <= 239:
points+=7
if 240 <= total_cholesterol <= 279:
points+=9
if total_cholesterol > 289:
points+=11
if 40 <= age <= 49:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=3
if 200 <= total_cholesterol <= 239:
points+=5
if 240 <= total_cholesterol <= 279:
points+=6
if total_cholesterol > 289:
points+=8
if 50 <= age <= 59:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=2
if 200 <= total_cholesterol <= 239:
points+=3
if 240 <= total_cholesterol <= 279:
points+=4
if total_cholesterol > 289:
points+=5
if 60 <= age <= 69:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=1
if 200 <= total_cholesterol <= 239:
points+=1
if 240 <= total_cholesterol <= 279:
points+=2
if total_cholesterol > 289:
points+=3
if 70 <= age <= 79:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=0
if 200 <= total_cholesterol <= 239:
points+=0
if 240 <= total_cholesterol <= 279:
points+=1
if total_cholesterol > 289:
points+=1
#smoking - male
if smoker:
if 20 <= age <= 39:
points+=8
if 40 <= age <= 49:
points+=5
if 50 <= age <= 59:
points+=3
if 60 <= age <= 69:
points+=1
if 70 <= age <= 79:
points+=1
else: # nonsmoker
points += 0
#hdl cholesterol
if hdl_cholesterol > 60:
points-=1
if 50 <= hdl_cholesterol <= 59:
points+=0
if 40 <= hdl_cholesterol <= 49:
points+=1
if hdl_cholesterol < 40:
points+=2
#systolic blood pressure
if not blood_pressure_med_treatment:
if systolic_blood_pressure < 120:
points+=0
if 120 <= systolic_blood_pressure <= 129:
points+=0
if 130 <= systolic_blood_pressure <= 139:
points+=1
if 140 <= systolic_blood_pressure <= 159:
points+=1
if systolic_blood_pressure >= 160:
points +=2
else: #if the patient is on blood pressure meds
if systolic_blood_pressure < 120:
points+=0
if 120 <= systolic_blood_pressure <= 129:
points+=1
if 130 <= systolic_blood_pressure <= 139:
points+=1
if 140 <= systolic_blood_pressure <= 159:
points+=2
if systolic_blood_pressure >= 160:
points +=3
#calulate % risk for males
if points <= 0:
percent_risk ="<1%"
elif points == 1:
percent_risk ="1%"
elif points == 2:
percent_risk ="1%"
elif points == 3:
percent_risk ="1%"
elif points == 4:
percent_risk ="1%"
elif points == 5:
percent_risk ="2%"
elif points == 6:
percent_risk ="2%"
elif points == 7:
percent_risk ="2%"
elif points == 8:
percent_risk ="2%"
elif points == 9:
percent_risk ="5%"
elif points == 10:
percent_risk ="6%"
elif points == 11:
percent_risk ="8%"
elif points == 12:
percent_risk ="10%"
elif points == 13:
percent_risk ="12%"
elif points == 14:
percent_risk ="16%"
elif points == 15:
percent_risk ="20%"
elif points == 16:
percent_risk ="25%"
elif points >= 17:
percent_risk =">30%"
#process females ----------------------------------------------------------
elif sex.lower()=="female":
# Age - female
if 20 <= age <= 34:
points-=7
if 35 <= age <= 39:
points-=3
if 40 <= age <= 44:
points-=0
if 45 <= age <= 49:
points+=3
if 50 <= age <= 54:
points+=6
if 55 <= age <= 59:
points+=8
if 60 <= age <= 64:
points+=10
if 65 <= age <= 69:
points+=12
if 70 <= age <= 74:
points+=14
if 75 <= age <= 79:
points+=16
#Total cholesterol, mg/dL - Female ------------------------
if 20 <= age <= 39:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=4
if 200 <= total_cholesterol <= 239:
points+=8
if 240 <= total_cholesterol <= 279:
points+=11
if total_cholesterol > 289:
points+=13
if 40 <= age <= 49:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=3
if 200 <= total_cholesterol <= 239:
points+=6
if 240 <= total_cholesterol <= 279:
points+=8
if total_cholesterol > 289:
points+=10
if 50 <= age <= 59:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=2
if 200 <= total_cholesterol <= 239:
points+=4
if 240 <= total_cholesterol <= 279:
points+=5
if total_cholesterol > 289:
points+=7
if 60 <= age <= 69:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=1
if 200 <= total_cholesterol <= 239:
points+=2
if 240 <= total_cholesterol <= 279:
points+=3
if total_cholesterol > 289:
points+=4
if 70 <= age <= 79:
if total_cholesterol < 160:
points +=0
if 160 <= total_cholesterol <= 199:
points+=1
if 200 <= total_cholesterol <= 239:
points+=1
if 240 <= total_cholesterol <= 279:
points+=2
if total_cholesterol > 289:
points+=2
#smoking - female
if smoker:
if 20 <= age <= 39:
points+=9
if 40 <= age <= 49:
points+=7
if 50 <= age <= 59:
points+=4
if 60 <= age <= 69:
points+=2
if 70 <= age <= 79:
points+=1
else: #nonsmoker
points += 0
#hdl cholesterol - female
if hdl_cholesterol > 60:
points-=1
if 50 <= hdl_cholesterol <= 59:
points+=0
if 40 <= hdl_cholesterol <= 49:
points+=1
if hdl_cholesterol < 40:
points+=2
#systolic blood pressure
if not blood_pressure_med_treatment: #untreated
if systolic_blood_pressure < 120:
points+=0
if 120 <= systolic_blood_pressure <= 129:
points+=1
if 130 <= systolic_blood_pressure <= 139:
points+=2
if 140 <= systolic_blood_pressure <= 159:
points+=3
if systolic_blood_pressure >= 160:
points +=4
else: #if the patient is on blood pressure meds
if systolic_blood_pressure < 120:
points+=0
if 120 <= systolic_blood_pressure <= 129:
points+=3
if 130 <= systolic_blood_pressure <= 139:
points+=4
if 140 <= systolic_blood_pressure <= 159:
points+=5
if systolic_blood_pressure >= 160:
points +=6
#calulate % risk for females
if points <= 9:
percent_risk ="<1%"
elif 9 <= points <= 12:
percent_risk ="1%"
elif 13 <= points <= 14:
percent_risk ="2%"
elif points == 15:
percent_risk ="3%"
elif points == 16:
percent_risk ="4%"
elif points == 17:
percent_risk ="5%"
elif points == 18:
percent_risk ="6%"
elif points == 19:
percent_risk ="8%"
elif points == 20:
percent_risk ="11%"
elif points == 21:
percent_risk ="14%"
elif points == 22:
percent_risk ="17%"
elif points == 23:
percent_risk ="22%"
elif points == 24:
percent_risk ="27%"
elif points >= 25:
percent_risk ="30%"
if errors:
response['status']=422
response['message'] = "The request contained errors and was unable to process."
response['errors']=errors
else:
response['points']=points
response['percent_risk']= percent_risk
return response
if __name__ == "__main__":
"""
Accept values to calucluate the 10 years heart attack risk based on
Framingham.
"""
try:
sex=sys.argv[1].lower()
age=sys.argv[2]
total_cholesterol=sys.argv[3]
hdl_cholesterol=sys.argv[4]
systolic_blood_pressure=sys.argv[5]
smoker=sys.argv[6].lower()
blood_pressure_med_treatment=sys.argv[7].lower()
except(IndexError):
print "All values are required."
print "Usage: framingham.py <sex> <age> <total_cholesterol> <hdl_cholesterol systolic_blood_pressure> <smoker> <blood_pressure_med_treatment>"
print "Example: framingham.py male 25 152 56 130 0 0"
exit(1)
try:
#execute the function
result = framingham_10year_risk(sex, age, total_cholesterol,
hdl_cholesterol, systolic_blood_pressure,
smoker, blood_pressure_med_treatment)
#return pretty-print json to standard out
print json.dumps(result, indent=4)
except():
print "An unexpected error occured. Here is the post-mortem:"
print sys.exc_info()