Mycobacterium tuberculosis is the primary bacteria responsible for tuberculosis, one of the most dangerous diseases, and top causes of death worldwide, as identified by the World Health Organization in a 2018 report. Diagnostic tools currently exist for identifying those who carry active or latent versions of the infection including chest radiographs, a Mantoux tuberculin skin test, or an acid-fast bacilli smear of sputum samples. These methods are standard in the medical community of high income countries, but pose challenges for lower-income regions of the world as well as vulnerable populations. The need for a rapid, inexpensive, and non-invasive method of tuberculosis detection is evident by the ten million that contracted and 1.6 million that died from TB in 2017 alone. In our study, we used a previously developed nanoplasmon-enhanced scattering technology combined with dark field microscopy in order to investigate the potential for a blood-based TB detection assay. Twenty-eight capture antibodies were screened using cell culture exosomes and human serum samples to identify candidates for a TB-derived exosome biomarker. Four antibodies demonstrated potential for distinguishing negative controls from positive controls in this study: anti-AG85, anti-AG85B, anti-SodA, anti-Ald. These capture antibodies displayed significant differences (p<0.05) for both cell culture exosomes and human serum samples on more than one occasion. The work is significant in its ability to distinguish potential capture antibody targets, and future experimentation may yield a technology ready for clinical settings to address the gap in current TB detection methods.