For every breast cancer detected by screening mammography, approximately three patients undergo biopsy. Two of the three biopsies will be benign, resulting in unnecessary biopsies particularly in women with dense breast tissue. Self-referencing differential spectroscopic analysis (SRDS) is a promising noninvasive optical imaging approach that uses functional over anatomic characteristics of the tissue to differentiate benign from malignant lesions. This method utilizes high-resolution spectroscopy to quantify the abundance of tumor-specific molecules that are not present in normal breast tissue. Diffuse optical spectroscopic imaging (DOI) has shown promise in performing SRDS in patients to reveal tumor-specific absorption signals, also known as "specific tumor component" (STC). However, the origin of the STC spectra remains unknown. Understanding the source of the STC spectrum is vital to improve the STC detection and ultimately achieve higher diagnostic performance. This research comprises two major objectives. One is to determine the meaning of the tumor specific absorption signature and understand its relation to the underlying tumor containing tissue absorption. The other objective is to deduce methods to improve the detection of the tumor specific absorption spectra as well as improve the quantification of major tissue constituents (blood, water and fat), and thereby improve breast cancer diagnosis. In order to accomplish these objectives, we have developed a new hyperspectral hybrid frequency domain (FD) and continuous wave (CW) diffuse optical imaging (hybrid DOI) technique, implemented with a simple probe configuration, to accurately measure 3D broadband optical properties of heterogeneous turbid media without having to employ spectral constraints. We simulated the reconstruction of absorption and scattering spectra (650 – 1000 nm) of human breast tumors in a homogeneous background at depths of 0 to 10 mm. The hybrid DOI technique demonstrated enhanced performance in reconstruction of optical absorption with a mean accuracy over all 71 wavelengths of 8.39 % versus 32.26 % for a 10 mm deep tumor with the topographic DOI method. The new hybrid technique was also tested and validated on two heterogeneous tissue-simulating phantoms with inclusion depths of 2, 7, and 9 mm. The mean optical absorption accuracy over all wavelengths was similarly improved up to 5x for the hybrid DOI method versus topographic DOI for the deepest inclusions.The SRDS method was evaluated by extracting the STC spectral absorption from the broadband absorption spectra retrieved via hybrid and conventional topographic DOI techniques on both computationally modelled and phantom measured data. The computational modelled and tissue-simulating phantom STC results were compared with the known underlying tumor/inclusion specific absorption spectra to reveal that the SRDS approach cannot establish quantitative/qualitative accuracy of the true tumor specific spectral absorption, which provided a new understanding about the SRDS method that has not been previously reported. Indeed, the STC content indicates more of a spectral misfit between the basis chromophore (blood, water and fat) contribution and the tumor and normal difference absorption, which implies the presence of other absorber/absorbers that are unaccounted for by the major tissue absorbers. Keeping these observations into account, we investigated the role of minor optical absorbers such as collagen and methemoglobin (metHb) in distinguishing benign and malignant breast lesions via noninvasive quantitative broadband DOI.A total of 28 subjects, including 12 benign and 18 malignant lesions (BIRADS score >= 3) were optically characterized in terms of oxy-, deoxy- hemoglobin, water, lipids, collagen, metHb and scattering parameters. Malignant lesions exhibited significantly higher lesion to normal ratios of oxy- , deoxy- hemoglobin, water, and metHb compared to benign lesions. Amongst all variables considered, only metHb was observed to be significant in lesion discrimination (p = 0.0016) without normalization to healthy tissue. Although the absolute concentration of metHb is small (0.43±0.18 µM for benign vs 0.87±0.32 µM for malignant), the significance of the differences is the strongest of all the parameters investigated. The high metHb concentration observed in malignant lesions compared to benign lesions reflects the angiogenic state of the lesions, and suggests variations in the composition of malignant tissue such as presence of neovascularization and hemorrhage. The discriminatory power of metHb for distinguishing malignant and benign lesions, also shows that this minor absorber was partially responsible for the greater magnitude of STC spectral content in malignant versus benign lesions, particularly below 665 nm and above 960 nm. The biochemical sources of the other STC spectral content (665 nm - 960 nm) are yet to be investigated and identified. Based on the results of this study, DOI-measured metHb shows significant potential in non-invasive lesion characterization and requires further investigation.In summary, a new hybrid DOI method was developed and validated for improved broadband characterization of deep tissue inhomogeneities using a simple reflectance-based probe. The SRDS method was evaluated on computational modelled and phantom measured broadband absorption to reveal a new understanding of the extracted STC spectral absorption. Methemoglobin was identified as one of the biochemical sources of STC spectral contrast between benign and malignant lesions, and for the first time, was identified as a potential optical biomarker of breast malignancy.