Update app.py

app.py

@@ -1,197 +1,52 @@
-import streamlit as st
-import numpy as np
-import random
 import plotly.graph_objects as go
 
-# Initialize agents (doctors, nurses, clinicians, patients)
-agents = {
-    "Doctors": 5,
-    "Nurses": 10,
-    "Clinicians": 8,
-    "Patients": 200
-}
-
-# Define states for patients (health conditions)
-patient_conditions = ["Healthy", "Mild Illness", "Chronic Illness", "Emergency"]
-
-# Define emotions for healthcare agents (stress, focus, etc.)
-doctor_emotions = ["Calm", "Stressed", "Overwhelmed"]
-nurse_emotions = ["Focused", "Fatigued", "Panicked"]
-
-# Define the actions doctors can take
-doctor_actions = ["Prescribe Medication", "Recommend Tests", "Consult Clinician", "Schedule Surgery"]
-
-# Define the actions nurses can take
-nurse_actions = ["Monitor Vitals", "Administer Medication", "Report to Doctor", "Assist Surgery"]
-
-# Reward and penalty system
-rewards = {
-    "Prescribe Medication": 12,  # Slightly increased reward
-    "Recommend Tests": 7,  # Slightly increased reward
-    "Consult Clinician": 9,
-    "Schedule Surgery": 17,
-    "Monitor Vitals": 4,
-    "Administer Medication": 12,
-    "Review Diagnostic Test": 7,
-    "Recommend Additional Tests": 6
-}
-
-penalties = {
-    "Wrong Medication": -5,  # Reduced penalty
-    "Missed Diagnosis": -10,
-    "Incorrect Test Recommendation": -3,
-    "Stress-Induced Mistake": -7
-}
-
-# Initialize the session state for counting treatments
-if "performance_metrics" not in st.session_state:
-    st.session_state.performance_metrics = {
-        "Doctor": {"successful_treatments": 0, "failed_treatments": 0},
-        "Nurse": {"successful_assists": 0, "failed_assists": 0}
-    }
-if "patient_satisfaction" not in st.session_state:
-    st.session_state.patient_satisfaction = 100
-
-# Define Reinforcement Learning algorithm (basic Q-learning approach)
-class Agent:
-    def __init__(self, agent_type):
-        self.agent_type = agent_type
-        self.q_table = np.zeros((len(patient_conditions), len(doctor_actions if agent_type == "Doctor" else nurse_actions)))
-        self.state = random.choice(patient_conditions)
-    
-    def choose_action(self, exploration_rate=0.03):  # Further reduce exploration rate
-        if random.uniform(0, 1) < exploration_rate:  # Explore
-            return random.randint(0, len(doctor_actions)-1 if self.agent_type == "Doctor" else len(nurse_actions)-1)
-        else:  # Exploit (choose the best action based on Q-values)
-            return np.argmax(self.q_table)
-    
-    def update_q_value(self, state, action, reward, learning_rate=0.25, discount_factor=0.95):  # Increase learning rate
-        old_q_value = self.q_table[state, action]
-        best_future_q_value = np.max(self.q_table)
-        new_q_value = old_q_value + learning_rate * (reward + discount_factor * best_future_q_value - old_q_value)
-        self.q_table[state, action] = new_q_value
-
-# Instantiate agents
-doctor_agent = Agent("Doctor")
-nurse_agent = Agent("Nurse")
-
-# Animation function
 def create_animation(successful_treatments, failed_treatments):
+    # Create initial figure
     fig = go.Figure()
 
-    # Add traces for successful and failed treatments
-    fig.add_trace(go.Bar(
-        x=['Successful Treatments', 'Failed Treatments'],
-        y=[successful_treatments, failed_treatments],
-        marker_color=['green', 'red'],
-        name='Doctor Performance'
-    ))
-
-    # Animation settings
+    # Add bar traces
+    fig.add_trace(go.Bar(x=['Doctor', 'Nurse'],
+                         y=[successful_treatments['Doctor'], successful_treatments['Nurse']],
+                         name='Successful Treatments',
+                         marker_color='green'))
+    
+    fig.add_trace(go.Bar(x=['Doctor', 'Nurse'],
+                         y=[failed_treatments['Doctor'], failed_treatments['Nurse']],
+                         name='Failed Treatments',
+                         marker_color='red'))
+
+    # Add animation frames (if needed for more dynamic changes over time)
+    frames = [go.Frame(data=[go.Bar(x=['Doctor', 'Nurse'],
+                                    y=[s, f])],
+                       name=f'Frame {i}') 
+              for i, (s, f) in enumerate(zip(successful_treatments.values(), failed_treatments.values()))]
+
+    # Set the layout for the plot
     fig.update_layout(
-        title="Doctor Performance Metrics",
-        xaxis_title="Treatment Type",
+        title="Doctor and Nurse Performance Metrics",
+        barmode='group',
+        xaxis_title="Personnel",
         yaxis_title="Number of Treatments",
-        updatemenus=[dict(type="buttons",
-                          buttons=[dict(label="Play",
-                                        method="animate",
-                                        args=[None, {"frame": {"duration": 500, "redraw": True}, "fromcurrent": True}])])],
-        frames=[go.Frame(data=[go.Bar(y=[successful_treatments + i, failed_treatments + i])], name=str(i))
-                for i in range(1, 10)]  # Example of dynamic range
+        updatemenus=[{
+            'buttons': [
+                {'args': [None, {'frame': {'duration': 500, 'redraw': True}, 'fromcurrent': True}], 
+                 'label': 'Play',
+                 'method': 'animate'},
+                {'args': [[None], {'frame': {'duration': 0, 'redraw': True}, 'mode': 'immediate'}], 
+                 'label': 'Pause',
+                 'method': 'animate'}
+            ],
+            'type': 'buttons'
+        }]
     )
 
-    return fig
-
-# Button to run the simulation
-if st.button("Run Simulation"):
-    # Simulate special event (disease outbreak or resource shortage)
-    special_event = random.choice([None, "Disease Outbreak", "Resource Shortage"])
-    if special_event:
-        st.subheader(f"Special Event: {special_event}")
-        if special_event == "Disease Outbreak":
-            st.write("A sudden disease outbreak has flooded the hospital with new patients. Resources are limited!")
-        elif special_event == "Resource Shortage":
-            st.write("A medical supply shortage is impacting the hospital. Staff must prioritize high-risk patients.")
-
-    # Patient Condition Simulation
-    patient_state = random.choice(patient_conditions)
-    st.write(f"Simulated Patient Condition: {patient_state}")
-
-    # Random doctor and nurse emotions (affects performance)
-    doctor_emotion = random.choice(doctor_emotions)
-    nurse_emotion = random.choice(nurse_emotions)
-    st.write(f"Doctor's Emotional State: {doctor_emotion}")
-    st.write(f"Nurse's Emotional State: {nurse_emotion}")
-
-    # Doctor Action Simulation
-    doctor_action = doctor_agent.choose_action()
-    st.write(f"Doctor's Chosen Action: {doctor_actions[doctor_action]}")
-
-    # Nurse Action Simulation
-    nurse_action = nurse_agent.choose_action()
-    st.write(f"Nurse's Chosen Action: {nurse_actions[nurse_action]}")
-
-    # Complications that can arise during treatment
-    complication = random.choices(
-        [None, "Allergic Reaction", "Unexpected Complication"],
-        weights=[0.6, 0.2, 0.2]
-    )[0]
-    
-    if complication:
-        st.subheader(f"Complication: {complication}")
-        if complication == "Allergic Reaction":
-            st.write("The patient has developed an allergic reaction to the prescribed medication!")
-            penalty = penalties.get("Wrong Medication", 0)
-            st.session_state.patient_satisfaction -= 10
-        elif complication == "Unexpected Complication":
-            st.write("An unexpected complication occurred during surgery!")
-            penalty = penalties.get("Stress-Induced Mistake", 0)
-            st.session_state.patient_satisfaction -= 15
-    else:
-        penalty = 0
-
-    # Ensure action constraints for healthy patients
-    if patient_state == "Healthy" and doctor_action == 0:  # 0 corresponds to "Prescribe Medication"
-        penalty = penalties.get("Wrong Medication", 0)
-        st.write("Prescribing medication to a healthy patient is unnecessary! Immediate penalty applied.")
-        st.session_state.patient_satisfaction -= 20
+    # Add frames to the figure for the animation
+    fig.frames = frames
 
-    # Reward or penalty based on action and emotional state
-    reward = rewards.get(doctor_actions[doctor_action], 0)
-    if doctor_emotion in ["Stressed", "Overwhelmed"]:
-        penalty += penalties.get("Stress-Induced Mistake", 0)
-
-    # Update Q-values
-    doctor_agent.update_q_value(patient_conditions.index(patient_state), doctor_action, reward if penalty == 0 else penalty)
-
-    st.write(f"Doctor's Reward: {reward if penalty == 0 else penalty}")
-    st.write(f"Patient Satisfaction: {st.session_state.patient_satisfaction}")
-
-    # Simulated patient feedback and next steps
-    next_step = random.choices(
-        ["Recovery", "Further Treatment Needed", "Complication"],
-        weights=[0.6, 0.3, 0.1]
-    )[0]
-    st.write(f"Patient Status after treatment: {next_step}")
-
-    # Increment performance based on outcome
-    if next_step == "Recovery":
-        st.session_state.performance_metrics["Doctor"]["successful_treatments"] += 1
-        st.session_state.performance_metrics["Nurse"]["successful_assists"] += 1
-    else:
-        st.session_state.performance_metrics["Doctor"]["failed_treatments"] += 1
-        st.session_state.performance_metrics["Nurse"]["failed_assists"] += 1
-
-    # Create animation for performance metrics
-    successful_treatments = st.session_state.performance_metrics["Doctor"]["successful_treatments"]
-    failed_treatments = st.session_state.performance_metrics["Doctor"]["failed_treatments"]
-
-    fig = create_animation(successful_treatments, failed_treatments)
-    st.plotly_chart(fig)
+    return fig
 
-# Display performance metrics
-st.subheader("Performance Metrics:")
-st.write(f"Doctor's Successful Treatments: {st.session_state.performance_metrics['Doctor']['successful_treatments']}")
-st.write(f"Doctor's Failed Treatments: {st.session_state.performance_metrics['Doctor']['failed_treatments']}")
-st.write(f"Nurse's Successful Assists: {st.session_state.performance_metrics['Nurse']['successful_assists']}")
-st.write(f"Nurse's Failed Assists: {st.session_state.performance_metrics['Nurse']['failed_assists']}")
+# Example call
+successful_treatments = {'Doctor': [10, 20, 30], 'Nurse': [5, 15, 25]}
+failed_treatments = {'Doctor': [2, 5, 10], 'Nurse': [1, 3, 6]}
+fig = create_animation(successful_treatments, failed_treatments)
+fig.show()