Spectroscopic needle probe for characterization of biological tissues

Biopsy is an invasive procedure in which a small piece of tissue is removed from the body to assess its properties and diagnose medical conditions such as inflammatory disorders, infections, immune diseases, or cancer. It is performed as prior and post treatment. Factors such as insufficient tissue volume, contamination (e.g., the presence of necrotic or fibrotic tissue), and sampling inadequacies can affect biopsy quality, potentially leading to delays in diagnosis. The presented thesis investigates the integration of machine-learning methods and advanced core needle biopsy (CNB) probes to improve the accuracy and adequacy of tissue sampling, complementing conventional guidance methods. The study focuses on the application of optical and impedance measurement methods for real-time detection of tissue alterations and classification of different tissues. Furthermore, the performance of classifiers is presented in model, ex vivo, and in vivo tissues. The findings underscore the potential of optical- and machine-learning-based methods to enhance CNB guidance, warranting further validation through expanded datasets and targeted studies on necrotic tissue characterization.