From 8536382e8b86db0c346ecf1d4b124d715471c71b Mon Sep 17 00:00:00 2001
From: Katie Zhuang <katie.zhi.zhuang@gmail.com>
Date: Thu, 2 Jun 2022 18:50:43 +0200
Subject: [PATCH] create some demo services for a neuro-cardiac example

---
 .../neuro_demo_services.py                    | 115 ++++++++++++++++++
 1 file changed, 115 insertions(+)
 create mode 100644 src/osparc_function_services/neuro_demo_services.py

diff --git a/src/osparc_function_services/neuro_demo_services.py b/src/osparc_function_services/neuro_demo_services.py
new file mode 100644
index 0000000..510d499
--- /dev/null
+++ b/src/osparc_function_services/neuro_demo_services.py
@@ -0,0 +1,115 @@
+
+from doctest import OutputChecker
+import random
+from numbers import Complex
+from typing import Tuple
+from enum import Enum
+
+from scipy import rand
+
+from pydantic import confloat, conint
+
+# Units validated against https://github.com/hgrecco/pint/blob/master/pint/default_en.txt
+
+
+######## This is a set of services intended for a "contextual" example of unit conversions ######
+# Concept:
+#   Heart function is affected by the autonomous nervous system via the parasympathetic (e.g. vagus)
+#   and sympathetic systems. We can stimulate the vagus nerve with electrical stimulation of a 
+#   particular amplitude and pulse width. 
+
+
+# Instructions:
+#   The pipeline is meant to be set up as follows:
+#   1. 2 float parameters for stimulation amplitude (mA) and pulse width inputs to stimulate autonomic_model
+#   2. 1 array float parameter for location of the center of electrode: dimension 1x3 (x,y,z coord)
+#   3. autonomic_model gets inputs from parameters 1 and 2. Outputs acetylcholine, epinephrine and norepinephrine 
+#       are connected to the inputs of sinoatrial_node. Output contractility is connected to cardiac_model.
+#   4. sinoatrial_node gets inputs acetylcholine, epinephrine and norepinephrine  from autonomic_model. The Output 
+#       heart rate is connected to the cardiac_model.
+#   5. cardiac_model receives one input from sinoatrial_node (heart rate) and one input from autonomic model 
+#       (contractility). It provides stroke volume and blood pressure as outputs, which can optionally be connected
+#       to the pressure_monitor. 
+#   6. pressure_monitor is an optional service that receives one input (blood pressure) from cardiac_model. 
+#       It will fail dynamic validation.
+#   
+# Pipeline: 
+#   electrode_loc_param --|
+#   stim_amp_param --------> autonomic_model -------------------------> cardiac_model --?--> pressure_monitor
+#   pulse_width_param-----|                  |---> sinoatrial_node---|
+#
+#
+# Note:
+#   Inputs and outputs are intentionally marked with different units (e.g. mM vs uM) to demonstrate conversions
+#
+
+
+class Exercise(str, Enum):
+    REST = "Resting"
+    LOW = "Low Activity"
+    MEDIUM = "Moderate Exercise"
+    HIGH = "Intense Exercise"
+
+
+# @services.add(return_outputs=["acetylcholine [mM]", "epinephrine [mM]", "norepinephrine [mM], "stroke_volume [mL]", "mean_arterial_pressure [kPa]"])
+def autonomic_model(
+    *,
+    stimulation_amp: confloat(ge=-1, le=1),                 # mA
+    pulse_width: confloat(ge=0, multiple_of=0.1, le=50),    # ms 
+    stim_location: list[confloat()],                        # list with units mm, of length==3 (x,y,z coordinate)
+    exercise_level: Exercise = Exercise.REST.value          # string
+) -> Tuple[float, float, float, float, float]:
+
+    options = {
+        Exercise.REST: 0.0,
+        Exercise.LOW: 0.25,
+        Exercise.MEDIUM: 0.5,
+        Exercise.HIGH: 0.75
+        }
+    acetylcholine = abs(stimulation_amp)            # mM (molar is a unit of concentration)
+    stroke_volume = random.randrange(40, 130)       # mL
+    mean_arterial_pressure = random.randrange(12, 20) # kPa 
+    epinephrine = options[exercise_level]           # mM 
+    norepinephrine = options[exercise_level] +.25   # mM
+
+    return acetylcholine, epinephrine, norepinephrine, stroke_volume, mean_arterial_pressure
+
+
+# @services.add(return_outputs=["heart_rate [bpm or Hz]"])
+def sinoatrial_node(
+    *,
+    acetylcholine: confloat(ge=0),  # uM (molar M is a unit of concentration)
+    epinephrine: confloat(ge=0),    # uM 
+    norepinephrine: confloat(ge=0)  # uM
+) -> int:
+
+    heart_rate = random.randint(0.5, 10)  # Hz
+
+    return heart_rate 
+
+# @services.add(return_outputs=["cardiac_output [L/min]", "blood_pressure [mmHg, mmHg]"])
+def cardiac_model(
+    *,
+    heart_rate: conint(ge=40, le=200),                  # rpm (would prefer bpm but not in pint)
+    stroke_volume: confloat(ge=0.004, le=0.130),        # L
+    mean_arterial_pressure: confloat(ge=90, le= 151)    # mmHg
+)-> Tuple[float, list]:
+
+    factor = random.random()
+    cardiac_output = factor*(20 - 4) + 4        # set output 4-20 L/min
+    sys_blood_pressure = factor*(140-100) + 100 # set systolic blood pressure 100-140 mmHg
+    dia_blood_pressure = factor*(90-70) + 70    # set diastolic blood pressure 70-90 mmHg
+    pressure = [dia_blood_pressure, sys_blood_pressure]
+
+    return(cardiac_output , pressure)
+
+
+## this is an optional service that will fail during runtime when connected to the output of cardiac_model
+# @services.add(return_outputs=["flag"])
+def pressure_monitor(
+    *,
+    pressures: list[confloat(le=90)]    # array with unit Pa (also pressure)
+)-> bool:
+
+    return True
+