Passively scanned, single-fiber optical coherence tomography probes for gastrointestinal devices.

BACKGROUND/OBJECTIVES: Optical coherence tomography (OCT) uses low coherence interferometry to obtain depth-resolved tissue reflectivity profiles (M-mode) and transverse beam scanning to create images of two-dimensional tissue morphology (B-mode). Endoscopic OCT imaging probes typically employ proximal or distal mechanical beam scanning mechanisms that increase cost, complexity, and size. Here, we demonstrate in the gastrointestinal (GI) tracts of unsedated human patients, that a passive, single-fiber probe can be used to guide device placement, conduct device-tissue physical contact sensing, and obtain two-dimensional OCT images via M-to-B-mode conversion. MATERIALS AND METHODS: We designed and developed ultrasmall, manually scannable, side- and forward-viewing single fiber-optic probes that can capture M-mode OCT data. Side-viewing M-mode OCT probes were incorporated into brush biopsy devices designed to harvest the microbiome and forward-viewing M-mode OCT probes were integrated into devices that measure intestinal potential difference (IPD). The M-mode OCT probe-coupled devices were utilized in the GI tract in six unsedated patients in vivo. M-mode data were converted into B-mode images using an M-to-B-mode conversion algorithm. The effectiveness of physical contact sensing by the M-mode OCT probes was assessed by comparing the variances of the IPD values when the probe was in physical contact with the tissue versus when it was not. The capacity of forward- and side-viewing M-mode OCT probes to produce high-quality B-mode images was compared by computing the percentages of the M-to-B-mode images that showed close contact between the probe and the luminal surface. Passively scanned M-to-B-mode images were qualitatively compared to B-mode images obtained by mechanical scanning OCT tethered capsule endomicroscopy (TCE) imaging devices. RESULTS: The incorporation of M-mode OCT probes in these nonendoscopic GI devices safely and effectively enabled M-mode OCT imaging, facilitating real-time device placement guidance and contact sensing in vivo. Results showed that M-mode OCT contact sensing improved the variance of IPD measurements threefold and side-viewing probes increased M-to-B-mode image visibility by 10%. Images of the esophagus, stomach, and duodenum generated by the passively scanned probes and M-to-B-mode conversion were qualitatively superior to B-mode images obtained by mechanically scanning OCT TCE devices. CONCLUSION: These results show that passive, single optical fiber OCT probes can be effectively utilized for nonendoscopic device placement guidance, device contact sensing, and two-dimensional morphologic imaging in the human GI tract in vivo. Due to their small size, lower cost, and reduced complexity, these M-mode OCT probes may provide an easier avenue for the incorporation of OCT functionality into endoscopic/nonendoscopic devices.

Mosaic Privacy-Preserving Mechanisms for Healthcare Analytics.

The Internet of Things (IoT) has propelled the evolution of medical sensing technologies to greater heights. Thus, traditional health systems have been transformed into new data-rich environments. This provides an unprecedented opportunity to develop new analytical methods and tools towards a new paradigm of smart and interconnected health systems. Nevertheless, there are risks pertinent to increasing levels of system connectivity and data accessibility. Cyber-attacks become more prevalent and complex, leading to greater likelihood of data breaches. These events bring sudden disruptions to routine operations and cause the loss of billions of dollars. Adversaries often attempt to leverage models to learn a target's sensitive attributes or extrapolate its inclusion within a database. As healthcare systems are critical to improving the wellbeing of our society, there is an urgent need to protect the privacy of patients and minimize the risk of model inversion attacks. This paper presents a new approach, named Mosaic Gradient Perturbation (MGP), to preserve privacy in the framework of predictive modeling, which meets the requirement of differential privacy while mitigating the risk of model inversion. MGP is flexible in fine-tuning the trade-offs between model performance and attack accuracy while being highly scalable for large-scale computing. Experimental results show that the proposed MGP method improves upon traditional gradient perturbation to mitigate the risk of model inversion while offering greater preservation of model accuracy. The MGP technique shows strong potential to circumvent paramount costs due to privacy breaches while maintaining the quality of existing decision-support systems, thereby ushering in a privacy-preserving smart health system.

Minimally Invasive Image-Guided Gut Transport Function Measurement Probe.

Introduction: Diseases such as celiac disease, environmental enteric dysfunction, infectious gastroenteritis, type II diabetes and inflammatory bowel disease are associated with increased gut permeability. Dual sugar absorption tests, such as the lactulose to rhamnose ratio (L:R) test, are the current standard for measuring gut permeability. Although easy to administer in adults, the L:R test has a number of drawbacks. These include an inability to assess for spatial heterogeneity in gut permeability that may distinguish different disease severity or pathology, additional sample collection for immunoassays, and challenges in carrying out the test in certain populations such as infants and small children. Here, we demonstrate a minimally invasive probe for real-time localized gut permeability evaluation through gut potential difference (GPD) measurement. Materials and Methods: The probe has an outer diameter of 1.2 mm diameter and can be deployed in the gut of unsedated subjects via a transnasal introduction tube (TNIT) that is akin to an intestinal feeding tube. The GPD probe consists of an Ag/AgCl electrode, an optical probe and a perfusion channel all housed within a transparent sheath. Lactated Ringer's (LR) solution is pumped through the perfusion channel to provide ionic contact between the electrodes and the gut lining. The optical probe captures non-scanning (M-mode) OCT images to confirm electrode contact with the gut lining. A separate skin patch probe is placed over an abraded skin area to provide reference for the GPD measurements. Swine studies were conducted to validate the GPD probe. GPD in the duodenum was modulated by perfusing 45 ml of 45 mM glucose. Results: GPD values of -13.1 ± 2.8 mV were measured in the duodenum across four swine studies. The change in GPD in the duodenum with the addition of glucose was -10.5 ± 2.4 mV (p < 0.001). M-mode OCT images provided electrode-tissue contact information, which was vital in ascertaining the probe's proximity to the gut mucosa. Conclusion: We developed and demonstrated a minimally invasive method for investigating gastrointestinal permeability consisting of an image guided GPD probe that can be used in unsedated subjects.

Gradient Mechanism to Preserve Differential Privacy and Deter Against Model Inversion Attacks in Healthcare Analytics.

Advanced sensing technologies, driven by the Internet of Things, have caused a sharp increase in data availability within the healthcare system. The newfound availability of data offers an unprecedented opportunity to develop new analytical methods to improve the quality of patient care. Data availability, however, is a double-edged sword. Malicious attacks and data breaches are increasingly seen in the healthcare field, which result in costly disruptions to operations. Adversaries exploit analytic models to infer participation in a dataset or estimate sensitivity attributes about a target patient. This paper is aimed at developing a differentially private gradient-based mechanism and assessing its utility in mitigating the impact of these attack risks within the context of the intensive care units. Experimental results showed that this methodology is capable of greatly reducing the risk of model inversion while retaining model accuracy. Thus, health systems that employ this technique can be given more peace of mind that high-quality services can be delivered in such a way that privacy is preserved.