deeprehab-diagnostics

Diagnostics functions for DeepRehab movement analysis results.

Installation

pip install deeprehab-diagnostics

Usage

Basic Example

from deeprehab_diagnostics import diagnose_movement

# Example angle data from deeprehab-angles package
left_angles = {
    "knee": 95.5,
    "shoulder": 165.2
}

right_angles = {
    "knee": 92.3,
    "shoulder": 162.1
}

# Angle series data for stability analysis
angle_series = [
    {"knee": 95.5, "shoulder": 165.2},
    {"knee": 94.8, "shoulder": 164.7},
    {"knee": 95.2, "shoulder": 165.0},
    {"knee": 94.9, "shoulder": 164.8}
]

# Perform comprehensive movement diagnostics
result = diagnose_movement("Deep Squat", left_angles, right_angles, angle_series)

print(f"Movement Type: {result.movement_type}")
print(f"Asymmetry Score: {result.asymmetry_score:.2f}")
print(f"Stability Score: {result.stability_score:.2f}")
print(f"Range of Motion: {result.range_of_motion}")
print(f"Recommendations: {result.recommendations}")
print(f"Risk Factors: {result.risk_factors}")

Squat Error Analysis

from deeprehab_diagnostics import analyze_squat_errors, generate_feedback

# Example angle data
angles = {
    "left_knee": 115,
    "right_knee": 135,
    "trunk_tilt": 25
}

# Analyze squat errors
errors = analyze_squat_errors(angles)
print(f"Detected errors: {errors}")

# Generate professional feedback
feedback = generate_feedback(errors)
print(f"Feedback: {feedback}")

Functions

diagnose_movement(movement_type, left_angles, right_angles, angle_series)

Perform comprehensive movement diagnostics.

Parameters:

• movement_type: Type of movement being analyzed
• left_angles: Dictionary of joint angles for left side
• right_angles: Dictionary of joint angles for right side
• angle_series: List of dictionaries containing angles for each frame

Returns:

• DiagnosticResult with comprehensive analysis

analyze_movement_symmetry(left_angles, right_angles)

Analyze symmetry between left and right side movements.

Parameters:

• left_angles: Dictionary of joint angles for left side
• right_angles: Dictionary of joint angles for right side

Returns:

• Asymmetry score (0-1, where 0 is perfectly symmetric)

analyze_movement_stability(angle_series)

Analyze stability of movement across frames.

Parameters:

• angle_series: List of dictionaries containing angles for each frame

Returns:

• Stability score (0-1, where 1 is perfectly stable)

generate_recommendations(asymmetry_score, stability_score, range_of_motion)

Generate recommendations based on diagnostic results.

Parameters:

• asymmetry_score: Movement asymmetry score
• stability_score: Movement stability score
• range_of_motion: Dictionary of joint ranges of motion

Returns:

• List of recommendations

identify_risk_factors(asymmetry_score, stability_score, range_of_motion)

Identify potential risk factors based on diagnostic results.

Parameters:

• asymmetry_score: Movement asymmetry score
• stability_score: Movement stability score
• range_of_motion: Dictionary of joint ranges of motion

Returns:

• List of identified risk factors

analyze_squat_errors(angles)

Analyze common errors in deep squat movement.

Parameters:

• angles: Dictionary containing joint angles and other measurements

Returns:

• Dictionary of detected errors

generate_feedback(errors)

Generate professional rehabilitation feedback based on detected errors.

Parameters:

• errors: Dictionary of detected errors from analyze_squat_errors

Returns:

• Professional, concise, and actionable feedback string

Integration with Other DeepRehab Packages

from deeprehab_pose import extract_landmarks
from deeprehab_angles import knee_angle, shoulder_angle
from deeprehab_diagnostics import diagnose_movement, analyze_squat_errors, generate_feedback

# Extract pose landmarks
landmarks_list = extract_landmarks("squat_video.mp4")

# Calculate angles for each frame
angle_series = []
left_angles_first_frame = {}
right_angles_first_frame = {}

for i, landmarks in enumerate(landmarks_list):
    left_knee = knee_angle(landmarks, "left")
    right_knee = knee_angle(landmarks, "right")
    left_shoulder = shoulder_angle(landmarks, "left")
    right_shoulder = shoulder_angle(landmarks, "right")
    
    frame_angles = {
        "knee": (left_knee + right_knee) / 2,
        "shoulder": (left_shoulder + right_shoulder) / 2
    }
    angle_series.append(frame_angles)
    
    # Save first frame angles for asymmetry analysis
    if i == 0:
        left_angles_first_frame = {
            "knee": left_knee,
            "shoulder": left_shoulder
        }
        right_angles_first_frame = {
            "knee": right_knee,
            "shoulder": right_shoulder
        }

# Perform comprehensive movement diagnostics
result = diagnose_movement(
    "Deep Squat", 
    left_angles_first_frame, 
    right_angles_first_frame, 
    angle_series
)

print(f"Movement Analysis Results:")
print(f"- Asymmetry Score: {result.asymmetry_score:.2f}")
print(f"- Stability Score: {result.stability_score:.2f}")
print(f"- Recommendations: {result.recommendations}")

# Analyze squat errors specifically
angles = {
    "left_knee": left_angles_first_frame["knee"],
    "right_knee": right_angles_first_frame["knee"]
}
errors = analyze_squat_errors(angles)
feedback = generate_feedback(errors)
print(f"Squat Error Feedback: {feedback}")

License

MIT