| Summary: | A new noninvasive and portable diagnostic system for detecting ocular tumors has been proposed. The system uses a contact-less electrically small antenna sensor to detect retinal cancer cells. The antenna sensor is operated in the ISM (Industrial, Scientific, and Medical) 2.413 GHz band and has electrical dimensions of 8 × 16.2 × 0.35 mm3. The antenna sensor is fabricated on a biodegradable Teslin substrate and tested in an eye-mimicking phantom to compare numerical computations with measurements. The specific absorption rate (SAR) obtained at near and far-field distances under 1 g of tissue is 1.18 W/kg and 0.353 W/kg, and that under 10 g of tissue is 0.112 W/kg and 0.313 W/kg, respectively. Furthermore, to detect the ocular tumor using the proposed antenna sensor, the resonance frequency shift, and the unsupervised machine learning technique, principle component analysis (PCA) is employed on simulated and measured results. The resonance frequency shift for a 3.5 mm radius tumor is 70 MHz for a single tumor and 120 MHz for double tumors. The PCA generates clusters with and without tumors on the positive and negative sides of the two-dimensional plot. The proposed techniques are more impactful in distinguishing between healthy and malignant tissues. The proposed systematic approach could be a portable platform for early detection of cancerous cells inside the eye.
|
|---|