Unavailability of large training datasets is a bottleneck that needs to be overcome to realize the true potential of deep learning in histopathology applications. Although slide digitization via whole slide imaging scanners has increased the speed of data acquisition, labeling of virtual slides requires a substantial time investment from pathologists. Eye gaze annotations have the potential to speed up the slide labeling process. This work explores the viability and timing comparisons of eye gaze labeling compared to conventional manual labeling for training object detectors. Challenges associated with gaze based labeling and methods to refine the coarse data annotations for subsequent object detection are also discussed. Results demonstrate that gaze tracking based labeling can save valuable pathologist time and delivers good performance when employed for training a deep object detector. Using the task of localization of Keratin Pearls in cases of oral squamous cell carcinoma as a test case, we compare the performance gap between deep object detectors trained using hand-labelled and gaze-labelled data. On average, compared to 'Bounding-box' based hand-labeling, gaze-labeling required 57.6% less time per label and compared to 'Freehand' labeling, gaze-labeling required on average 85% less time per label.

This paper presents the design of a new paradigm for a content-aware wireless MAC layer that is optimized for wireless video (first and foremost) while targeting fairness and stability among competing video traffic, and among video and non-video traffic. Hence, we refer to the proposed MAC framework as the FAST (Fair And STable) protocol. FAST employs two parameters for each packet, a quality value and a time-to-live value. Based on these parameters, FAST is designed on a multiclass priority queuing system that classifies the incoming traffic according to the content of each traffic flow and further identifies different priorities within each video content. We develop analytical frameworks to formulate channel allocation based on video/non-video fairness and video stability requirements as a joint bandwidth maximization and scheduling optimization problem. We incorporate these frameworks to design and simulate a content-aware channel access mechanism, which utilizes video traffic content classifications and users demand in conjunction with stability and fairness requirements at the MAC layer to allocate wireless channels to individual wireless users. Our simulation results show that FAST provides significant improvements in packet-loss-ratio, delay, overall fairness, and stability parameters when compared with leading access control mechanisms over 4G/LTE environment.