Smart Electrosurgical Knife For Real-Time Intraoperative Tissue Detection

Breast cancer is the most common cancer in women globally, and breast-conserving surgery (BCS), or lumpectomy, is the primary treatment for early-stage patients. However, achieving clear margins—when the tumor is fully removed—remains a challenge, often requiring re-excision surgeries or additional treatments. Surgeons rely on preoperative images and palpation to estimate the tumor’s borders, but these methods are not always accurate, with incomplete resections occurring in 10% to 50% of cases. In this thesis, we explore the integration of Diffuse Reflectance Spectroscopy (DRS), an optical tissue-sensing technology, into an electrosurgical knife to create a smart electrosurgical knife that addresses this issue. DRS helps to distinguish between cancerous and healthy tissues in real-time by analyzing optical properties, offering a potential solution to ensure clear margins during surgery. Researchers have shown that DRS can differentiate tissues based on their unique optical fingerprints, such as the Fat/Waterratio, which helps determine tumor borders.

We began by identifying the main challenges in combining DRS with an electrosurgical knife. Initial tests on porcine tissue examined the impact of electrosurgery on the optical fibers and their ability to distinguish between tissues (Chapter 2). Microscopic analyses revealed changes in the optical fibers and the formation of debris during electrosurgery, which could interfere with accurate tissue sensing. The chapter concludes with an assessment of the fibers' performance in delivering light and capturing DRS readouts before and after electrosurgery, highlighting the complexities of using this technology in a surgical environment and the need for design modifications to protect the optical fibers from the effects of electrosurgery.

To further investigate design iterations and verify the technology, we describe in Chapter 3 the developed tissue-mimicking phantom materials that replicate the optical properties of human breast tissue. Since breast tissue consists of multiple layers with varying optical characteristics, simulating these complexities was crucial to test the smart electrosurgical knife. These phantom materials mimic both healthy and cancerous tissues, providing a controlled environment to evaluate the knife’s ability to differentiate between tissue types. The results from this chapter helped refine both the design and testing methodology of the knife.

In Chapter 4, we focus on the continued development of the smart electrosurgical knife to ensure it can withstand the demands of prolonged surgery. Several designs were tested on porcine tissue to identify the most effective configuration for integrating DRS. These designs were evaluated for their ability to maintain accurate DRS readings while performing electrosurgery, which involves high temperatures and tissue coagulation. Further experiments using tissue-mimicking materials demonstrated the knife’s ability to identify distinct tissue layers in real-time during electrosurgery.

In Chapter 5, we continued to enhance the design process by incorporating feedback from clinicians. In the first part, the best-performing design from previous experiments was refined based on suggestions from surgeons who tested the device. Various design concepts were produced to optimize the knife's electrosurgical and cutting performance while ensuring the integration of DRS technology. In the second part, the knife's performance was evaluated during a simulated lumpectomy on a tumor-containing phantom. Surgeons performed the surgery four times—twice with a traditional electrosurgical knife relying on preoperative images, and twice with the smart electrosurgical knife providing real-time tissue sensing. The results showed that using the smart knife led to better surgical outcomes, highlighting its potential as an intraoperative tool for margin assessment and its promise for improving the precision and safety of breast cancer surgeries.

In Chapter 6, we focus on simplifying the technology for broader surgical use. We proposed using LEDs and photodetectors instead of wide-band light sources and spectrometers, creating a compact, console-free design. An electronic board was developed to identify optimal wavelengths for breast tissue, and a proof-of-concept showed that this system could distinguish tissue-mimicking materials. We envision a handheld device integrating DRS into the electrosurgical knife, removing the need for bulky equipment and paving the way for a more accessible, user-friendly tool in clinical settings.

Finally, Chapter 7 reviews the key findings from the research and presents concluding remarks. The thesis demonstrates the potential of integrating DRS into surgical tools for real-time tissue identification, offering a valuable solution to the problem of incomplete tumor resection.