Development of a Feedback Control System for Femtosecond Pulsed Laser Surgery.

摘要

There is a growing interest in ultra-short pulsed laser surgery combined with a feedback control mechanism for high-precision cutting with minimal collateral damage. Compared to conventional nanosecond lasers, ultra-short pulsed lasers do not deposit as much heat in the sample and therefore do not cause adverse thermal effects such as fractures. For implementation in the clinic, a real-time feedback control system will ensure that the laser is incident on the targeted tissue and does not cause damage to surrounding tissues. There are a variety of optical spectroscopy techniques that have the potential to be successfully incorporated into a feedback control mechanism for laser surgery. The technique investigated in this dissertation is the basic atomic spectroscopy technique, laser induced breakdown spectroscopy (LIBS), since it would allow the use of the same laser that is being used for ablation.;The main objective of this work is to investigate how the LIBS signal changes under different experimental conditions with implications towards using this technique as a feedback control system for ultra-short pulsed laser surgery applications. To achieve this objective, this dissertation work was split into three aims: characterizing the LIBS signal as a function of various experimental parameters, evaluating the effect of repetition rate on the LIBS signal, and assessing whether LIBS could be used to distinguish between tumor bone and normal bone. LIBS signals are produced when the laser fluence exceeds a given threshold for molecular breakdown of a sample. Immediately following, bound electrons are ionized and a plasma plume is formed. When the laser excitation source is turned off, the free electrons recombine with positively charged ions and atoms. Once these ions and atoms relax from their excited states, they emit characteristic atomic lines that can be detected. The ejected or evaporated material leads to crater formation on the sample.;To test the first aim, the LIBS intensity dependence on different parameters such as sample, laser energy, scanning speed, and depth of focus was characterized. The LIBS spectra were acquired using a custom-built upright microscope to tightly focus a femtosecond pulsed laser with an objective lens on to bone and soft tissue. The LIBS bone and soft tissue spectra showed distinct differences indicating that they can be successfully distinguished from one another. The bone LIBS spectrum showed several strong calcium peaks and a sodium peak in the spectral region we analyzed. The soft tissue LIBS spectrum showed a single distinct sodium peak near 590 nm. The LIBS signal for bone was also characterized as a function of the following excitation parameters: laser energy, depth of focus, and number of pulses per focal volume. A linear increase in the LIBS intensity as a function of the laser energy from 25 to 75 muJ was observed. In addition, we showed that moving the beam out of focus and the presence of overlapping pulses on the same focal area leads to a decrease in LIBS intensity due to changes in focal spot size. The LIBS intensity varied by an order of magnitude when the laser was moved out of focus. These results indicated that a potential feedback system for laser surgery should not be directly dependent on LIBS intensity. The feedback control algorithm should instead rely on a ratio algorithm that compares two or more unique spectral peaks. We compared the ratio between the calcium peak intensity at 612 nm to the sodium peak intensity at 589 nm. Under conditions where the repetition rate is constant at 1 kHz and there is a minimal temperature rise, the ratio between the calcium to sodium peak to be constant over a depth of approximately 1 mm.;For practical considerations in the clinic, it is not enough to have a real-time feedback control system for laser surgery. The femtosecond laser cutting speed needs to be increased to a level that is comparable to or faster than the mechanical tools currently used for hard tissue surgeries. For this reason, the second aim of the dissertation work focuses on evaluating the effect of increasing laser repetition rate on ablation width, sample temperature, and LIBS signal of bone. SEM images were acquired to quantify the morphology of the ablated volume and LIBS was performed to characterize changes in signal intensity and background. For the first time, experimentally measured temperature distributions of bone irradiated with femtosecond lasers at repetition rates below and above carbonization conditions are shown. At repetition rates where carbonization occurs, the sample temperature increases to a level that is well above the threshold for irreversible cellular damage. At these carbonizing repetition rates, we also observed a change in the LIBS spectrum. At repetition rates above carbonization, the calcium to sodium peak ratio for bone decreases to a value that is similar to soft tissue. Under these conditions, the bone would be misclassified as soft tissue using LIBS. These results highlight the importance of the need for careful selection of the repetition rate for a femtosecond laser surgery procedure to minimize the extent of thermal damage to surrounding tissues and prevent misclassification of tissue by LIBS analysis.;The third aim centers on assessing whether LIBS can also be used to distinguish between tumor and normal bone with the potential to be used to identify tumor margins in real-time. To test the third aim, preliminary LIBS measurements were performed on a primary bone tumor and matched normal bone sample. Analysis of the LIBS spectra showed a change in the magnesium peak intensity relative to the calcium peak intensity between primary bone tumor and normal bone. These results show the potential for LIBS to be used for identifying tumor margins in real-time to ensure complete removal of tumor tissue.;This dissertation demonstrates that LIBS shows promising potential to be used as a real-time feedback control system for ultra-short pulsed laser surgery procedures. The long-term vision for this work is to develop a fully integrated feedback control system for ultra-short pulsed laser ablation surgery to have high precision cutting with minimal damage to surrounding tissues. The feedback control system should divert the laser beam when it is incident on surrounding tissues based on their spectral signature. This would allow for high precision cutting with minimal damage to surrounding tissues. In the proposed case of spinal surgery, the feedback control system would ensure that the laser surgical tool only cuts bone and prevents damage to the spinal nerve.