Design and development of a machine learning enabled autofluorescence device for preoperative breast lump diagnosis and intraoperative breast tumor margin assessment

Adult female patients with a breast lump will be seen by a consultant surgeon in the Department of General Surgery, Kasturba Hospital, Manipal. She will get admitted if a surgical intervention is mandated as Standard of Care for the condition. The surgeon / clinician co-Investigator will obtain informed consent from the patient. Only patients consenting for the study, after she has been  explained in the language she understands, will be enrolled into the study. Patient and disease associated data as mentioned in the proforma will be obtained from the medical records of the patient

 For this first phase of the study, after the specimen is removed by the consultant as a part of standard of care, the novel device will be passed into the specimen (ex vivo) in the OT itself. Recording of spectra will be done and archived. After the recordings are completed the specimen will be sent to pathology lab, where the pathologist will bisect the specimen and further recordings of spectra will be done from the surface and within the lump. After all recordings are completed, the specimen will be processed by the pathology department for routine HPE. There is no additional blood or tissue sample harvested from the patient. There are no follow up visits.

After the spectral signatures are obtained, a machine learning algorithm (software) will be tested to improve the sensitivity, specificity, positive predictive value and negative predictive value of the device in real time in comparison to the HPE diagnosis