Tactile, Orientation, and Optical Sensor Fusion for Tactile Breast Image Mosaicking

Abstract

Breast cancer screening using Tactile Imaging (TI) is an advancing field of low-cost non-invasive medical imaging. Utilizing arrays of capacitive pressure transducers to perform a differential stress measurement of suspicious tissue, TI has been shown to be effective in measuring lesion size and stiffness, and subsequent differentiation of malignant and benign conditions, in repeated clinical studies. In order to further improve the lesion classification accuracy of clinical TI, this paper presents a novel method of mosaicking tactile images to form a large composite tactile map using the vein structure within the breast to spatially register tactile data. This paper demonstrates practical non-rigid tactile image mosaicking, using probe contact force and relative orientation sensor fusion to correct for the tissue deformation during tactile scanning, miniaturized and applied to a pre-clinical TI prototype. Testing of the proposed TI prototype on representative, tissuemimicking, silicone breast phantoms, with varying baseline elasticity and internal vein structure, yields typical image registration accuracies of 0.33% ± 0.15%. In similar testing, the proposed system measures the background elasticity of the samples with worst case error < 4.5% over the range 9 kPa to 60 kPa, required for accurate lesion characterization. This work will lead into further clinical validation of TI for measurement and classification of in-situ phantom and breast lesions, utilizing the delivered metrics from this work to improve differentiation accuracy.

Phantom Study of Arterial Localization using Tactile Sensor Array and a Normal Vs. Shear Pulse Pressure Propagation Method

“…This BP measurement methods can be applied to this array, but not vice-versa. The depth measurement is likely not to be practical clinically due to interference from surrounding bones and non-linear tissue, however it is expected to have applications in other clinical tactile imaging applications [16].…”

Using positional tracking to improve abdominal ultrasound machine learning classification

“…In order to test the efficacy of positional based tracking of an ultrasound probe for machine learning, two separate systems were tested: optical infrared tracking (IR) using a Vicon system, and an IR system based upon low-cost application specific integrated circuits (ASICs) IR sensors. Vicon has been shown to be highly accurate with within 2 mm [11], and is effective as a positional and registration reference measurement in other medical imaging applications [12,13]. It also has shown to achieve high accuracies in motion capture, as part of complex automated classification processes such as respiratory tracking [14] and pose estimation [15].…”

Steady State Physical Modeling for Optimizing Capacitive Tactile Sensors Thermal Sensitivity

“…APACITIVE tactile pressure sensors are a common method of measuring and quantifying tactile interactions, and are routinely employed in applications in contact with the human body [1], such as in medical imaging [2], [3], medical therapeutics [4]- [6], haptics and wearables [7], [8], and ergonomic testing [9] amongst others. These applications often involve large temperature variations and differentials during active operation, where performance is expected to be maintained throughout.…”