# Low-Cost Upper-Limb Visuomotor Adaptation & Synchronous Ne...

Prompt

# Low-Cost Upper-Limb Visuomotor Adaptation & Synchronous Neuromuscular Profiling Platform **Theme:** Frugal Neuro-Rehabilitation Diagnostics **Author:** MBBS Student Project Showcase **Target Venue:** Karachi Youth Summit (Expo Centre Karachi, Hall 6) --- ## 1. Executive Summary & Clinical Intent This platform is a low-cost, decentralized diagnostic and profiling rig designed to evaluate upper-limb sensorimotor adaptation and feedforward motor control in stroke hemiparesis and neuromotor deficits. Commercial motion capture (Vicon/Qualisys) and laboratory-grade electromyography (Biopac/Delsys) suites cost several millions of PKR and require specialized technicians. This project delivers synchronized kinematic and surface electromyographic (sEMG) profiling using an accessible BOM under $65 USD (~Rs. 17,500 PKR), making high-precision rehabilitation diagnostics deployable in district health facilities. --- ## 2. Core Extracted Metrics 1. **Initial Direction Error (IDE):** Angular deviation ($\theta_{\text{error}}$) between the vector connecting the reach origin to the actual target and the trajectory vector measured at early reach displacement (5–8 cm). Reflects feedforward motor planning prior to feedback correction. 2. **Peak Tangential Velocity ($v_{\text{peak}}$):** $$v(t) = \sqrt{\left(\frac{dx}{dt}\right)^2 + \left(\frac{dy}{dt}\right)^2}$$ Identifies motor trajectory smoothing, acceleration profiles, and sub-movements during reach execution. 3. **Electromechanical Delay (EMD):** $$\text{EMD} = t_{\text{movement\_onset}} - t_{\text{sEMG\_burst\_onset}}$$ Calculates the temporal lag (typically 30–50 ms in healthy subjects) between the electrical depolarization of the target muscle (*Brachioradialis* / *Triceps*) and the physical acceleration liftoff of the fingertip. --- ## 3. System Hardware Architecture * **Vision Tracking Sensor:** UGREEN CM831 / 65381 (8MP Sony IMX CMOS sensor) running at 1080p @ 60 FPS via MJPEG over USB Video Class (UVC). Locked optical focal plane (0 focus breathing) and locked manual shutter speed (0 dynamic exposure drift). * **Kinematic Plane:** $40 \times 30\text{ cm}$ white acrylic planar reaching workspace. Overhead gooseneck boom arm mounted via the camera's integrated 1/4"-20 brass tripod socket. * **Spatial Calibration:** 4-point planar homography transformation ($H \in \mathbb{R}^{3 \times 3}$) mapping raw pixel coordinates $[u, v, 1]^T$ directly into physical metric coordinates $[X, Y, 1]^T$ in millimeters ($0.32\text{ mm/pixel}$ resolution). * **Electrophysiology Front-End:** Single-channel analog differential sEMG circuit reading agonist/antagonist arm groups via disposable Ag/AgCl snap electrodes. * **ADC Digitization Bridge:** ADS1115 16-bit high-precision $I^2C$ ADC module (configured up to 860 SPS) interfaced to an Arduino Nano / ESP32 acting as a USB-serial packet bridge to Python. --- ## 4. Bill of Materials (BOM) & Karachi Procurement | Component | Exact Specification | Sourcing Location (Karachi) | Est. Cost (PKR) | | :--- | :--- | :--- | :--- | | **Overhead Optical Sensor** | UGREEN CM831 / SKU 65381 (1080p60, Sony IMX, 1/4" mount) | Computer Zone (Gulshan) / RB Tech | Rs. 8,999 – 9,800 | | **Mounting Boom Arm** | Articulated desk clamp arm with 1/4"-20 male post | Saddar Electronics / Camera Market | Rs. 1,800 – 2,500 | | **Analog sEMG Module** | Differential muscle sensor + 3-lead snap cable | Electronation PK / Saddar Regal | Rs. 2,200 – 3,000 | | **16-bit ADC** | ADS1115 4-channel $I^2C$ module | Regal Trade Square (2nd Floor, Saddar) | Rs. 750 – 900 | | **Serial MCU Bridge** | Arduino Nano (CH340G) or ESP32 | Regal Trade Square / Electronation PK | Rs. 750 – 1,400 | | **Bio-Electrodes** | Disposable pre-gelled Ag/AgCl snap electrodes (Pack of 30-50) | Pharmacy near Civil / Dow / LuckyOne | Rs. 600 – 900 | | **Task Board & Optics** | $40 \times 30\text{ cm}$ white acrylic board + matte neon-green stickers | Saddar acrylic shop / Urdu Bazar | Rs. 850 – 1,400 | | **TOTAL** | | | **~Rs. 16,000 – 19,500** | --- ## 5. Production Tracking & Initialization Code (Python / OpenCV) ```python import cv2 import numpy as np # 1. Initialize DirectShow capture backend cap = cv2.VideoCapture(0, cv2.CAP_DSHOW) # 2. Force MJPEG container to guarantee 1080p @ 60 FPS over USB bandwidth cap.set(cv2.CAP_PROP_FOURCC, cv2.VideoWriter_fourcc(*'MJPG')) cap.set(cv2.CAP_PROP_FRAME_WIDTH, 1920) cap.set(cv2.CAP_PROP_FRAME_HEIGHT, 1080) cap.set(cv2.CAP_PROP_FPS, 60) # 3. Hard-lock focus motor to eliminate focus breathing cap.set(cv2.CAP_PROP_AUTOFOCUS, 0) cap.set(cv2.CAP_PROP_FOCUS, 0) # 4. Hard-lock manual exposure to enforce Constant Frame Rate (CFR) cap.set(cv2.CAP_PROP_AUTO_EXPOSURE, 0.25) cap.set(cv2.CAP_PROP_EXPOSURE, -6) # (Optional: Pop native Windows driver dialogue to verify hardware registers) # cap.set(cv2.CAP_PROP_SETTINGS, 1) print("Optical sensor pipeline initialized at 1080p @ 60 FPS (Hardware Locked).")