Abstract
<jats:p>This chapter provides an engineering-oriented overview of spectral domain optical coherence tomography (SD-OCT), with particular emphasis on the device and system-level design choices that govern practical imaging performance. Rather than presenting a broad or purely descriptive survey, the chapter is structured around three closely related technical themes: broadband spectrometer design for achieving high axial resolution, k-linearization strategies for reducing resampling error and mitigating sensitivity roll-off, and balanced detection architectures for improving signal-to-noise ratio while suppressing direct current (DC) background and autocorrelation artifacts. System performance is discussed in terms of axial resolution, imaging depth, sensitivity roll-off, signal-to-noise ratio, dynamic range, and computational burden, and each metric is related to the underlying optical configuration, calibration procedure, and dominant sources of error. Representative examples, including several from our recent work, are used to illustrate how these design considerations are implemented in practice, including broadband ultra-high-resolution SD-OCT (UHR-SD-OCT) spectrometers, hardware-assisted k-linear sampling based on prism and freeform optical designs, and single spectrometer balanced detection SD-OCT. The chapter further examines the tradeoffs among resolution, depth range, robustness, and processing complexity, with the aim of providing practical guidance for the design, optimization, and implementation of high-performance SD-OCT systems for biomedical and related applications.</jats:p>