Unilateral breast cancer-related lymphedema diagnosis and monitoring system using bioimpedance analysis on a smartphone application / Syarifah Aisyah Syed Ibrahim

Syarifah Aisyah , Syed Ibrahim (2024) Unilateral breast cancer-related lymphedema diagnosis and monitoring system using bioimpedance analysis on a smartphone application / Syarifah Aisyah Syed Ibrahim. Masters thesis, Universiti Malaya.

	PDF (The Candidate's Agreement) Restricted to Repository staff only Download (104Kb)
	PDF (Thesis M.A.) Download (3638Kb)

Abstract

Lymphedema resulting from breast cancer treatment can significantly impact the quality of life for affected individuals. Early diagnosis and continuous monitoring of lymphedema are crucial for effective management. However, current methods are inconvenient, expensive, and limited accessibility. This thesis presents the development of a portable system for diagnosing and monitoring unilateral breast cancer-related lymphedema (UBCRL) using bioimpedance analysis (BIA) and a smartphone application. The developed UBRCL system utilises BIA technique via bioimpedance spectroscopy frequency analysis. The bioimpedance analyser includes an impedance analyser unit (AD5941), isolating capacitors (0.47 μF), limiting resistors (1 kΩ), electrostatic discharge protection unit, calibration resistors (1 kΩ), and ADuCM3029 microcontroller. The AD5941 embeds an AC voltage signal generator, high-precision current meter, differential voltage meter, and Discrete Fourier transform (DFT). The DFT processes measured data, calculating real and imaginary parts, while the ADuCM3029 microcontroller computes bioimpedance values. The system applies tetrapolar electrode configuration for uniform current distribution and eliminates skin contact impedance. Frequencies from 5 kHz to 200 kHz with 400 μA are delivered to the upper limbs through source electrodes. Resultant bioimpedance measurements are captured using sense electrodes. The bioimpedance measurements are wirelessly transmitted to a smartphone application via Bluetooth. Data is sent to the Google Cloud Function to extrapolate Cole parameters, specifically the resistance at zero frequency (R0), and evaluate inter-arm R0 ratio. The R0 ratio is stored in Firestore database and on the app, aiding UBCRL diagnosis and monitoring. The smartphone application was developed using the Flutter framework. The app functions to initiate measurements with the bioimpedance analyser and provides a user interface for diagnosing and monitoring UBCRL. The UBCRL system was tested and validated using passive loads and clinical validation involving 45 healthy participants and 100 unilateral BCRL patients. The results were compared with established devices: LCR meter and Quadscan 4000. The validation results using passive loads against an LCR meter demonstrated high accuracy and precision with 0.5% error, 0.01 standard deviation, 0.08% precision, and R–squared value of 0.999. Validation results compared with Quadscan 4000 showed negligible difference (p > 0.05), indicating high similarity with the commercial device. The clinical validation showed that the developed UBCRL system could differentiate between healthy and lymphedema-affected upper limbs. The R0 ratios were significantly different between healthy participants and stage 0 and those with stage 1 (p < 0.001) and stage 2 (p < 0.001) lymphedema. The UBCRL detection threshold in Malaysia was identified as 1.077 for dominant affected and 1.047 for non–dominant affected individuals. Mean and standard deviation dominant affected individuals were: stage 0 (1.034 ± 0.021), stage 1 (1.145 ± 0.035), stage 2 (1.394 ± 0.090). For non–dominant affected individuals: stage 0 (0.989 ± 0.029), stage 1 (0.989 ± 0.014), stage 2 (1.447 ± 0.712). These findings demonstrate the potential of the developed UBCRL system for diagnosis, monitoring and effective treatment management of UBCRL. The system is non-invasive, portable and affordable, increasing access particularly in resource-limited settings.

Actions (For repository staff only : Login required)