Dataset Information

Handheld Briefcase Optical Coherence Tomography with Real-Time Machine Learning Classifier for Middle Ear Infections.

ABSTRACT: A middle ear infection is a prevalent inflammatory disease most common in the pediatric population, and its financial burden remains substantial. Current diagnostic methods are highly subjective, relying on visual cues gathered by an otoscope. To address this shortcoming, optical coherence tomography (OCT) has been integrated into a handheld imaging probe. This system can non-invasively and quantitatively assess middle ear effusions and identify the presence of bacterial biofilms in the middle ear cavity during ear infections. Furthermore, the complete OCT system is housed in a standard briefcase to maximize its portability as a diagnostic device. Nonetheless, interpreting OCT images of the middle ear more often requires expertise in OCT as well as middle ear infections, making it difficult for an untrained user to operate the system as an accurate stand-alone diagnostic tool in clinical settings. Here, we present a briefcase OCT system implemented with a real-time machine learning platform for middle ear infections. A random forest-based classifier can categorize images based on the presence of middle ear effusions and biofilms. This study demonstrates that our briefcase OCT system coupled with machine learning can provide user-invariant classification results of middle ear conditions, which may greatly improve the utility of this technology for the diagnosis and management of middle ear infections.

SUBMITTER: Won J

PROVIDER: S-EPMC8147830 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PURPOSE:Central foveal thickness (CFT) measurements from optical coherence tomography (OCT) scans provide a precise measure of severity of pathologic changes in the fovea, progress of disease and response to treatment. Although these measures are additionally valuable to assess foveal development in infants, their reproducibility is not known. The goal of this retrospective study is to evaluate the variation and reproducibility of CFT measurements using handheld spectral-domain OCT (hh-SDOCT) in supine infants compared to conventional adult tabletop imaging. METHODS:Imaging sessions with multiple macular, volume scans in one eye were selected for analysis from two participant groups: Group 1, 25 imaging sessions from 21 preterm infants without macular edema imaged supine in the nursery using hh-SDOCT (Leica/Bioptigen Envisu C2300, RTP, NC); Group 2, 25 imaging sessions from 25 adults imaged using tabletop Bioptigen SDOCT. For each imaging session, three macular OCT volumes with acceptable image quality were selected for analysis. CFTs were measured using a customized script for automatic segmentation. An expert grader and a typical grader corrected the segmentation lines for the central foveal frame. Coefficient of variations (CV) and intraclass correlation coefficients (ICC) were calculated for graders and systems and compared to the previous literature on OCT reproducibility. RESULTS:CFT measurements were repeatable and reproducible for both handheld and tabletop SDOCT systems. For handheld, grader ICC (CI) and mean CV were 0.94 (0.90-0.97) and 3.8 (typical) and 0.98 (0.96-0.99) and 2.9 (expert), and for tabletop were 0.91(0.83-0.96) and 2.1 (typical) and 0.92 (0.86-0.96) and 1.9 (expert). Intergrader reproducibility of handheld and tabletop SDOCT systems were ICC(CI) 0.97 (0.95-0.98) and 0.93 (0.89-0.96) respectively, and both are comparable to previously reported reproducibility of tabletop systems. CONCLUSION:Handheld SDOCT is a reproducible instrument to measure foveal thicknesses in supine infants. It can be used in clinical research to evaluate foveal changes during retinal development and pathological conditions.

Project description:PurposeTo demonstrate the feasibility of a miniature handheld optical coherence tomography (OCT) imager for real time intraoperative vascular patency evaluation in the setting of super-microsurgical vessel anastomosis.MethodsA novel handheld imager Fourier domain Doppler optical coherence tomography based on a 1.3-µm central wavelength swept source for extravascular imaging was developed. The imager was minimized through the adoption of a 2.4-mm diameter microelectromechanical systems (MEMS) scanning mirror, additionally a 12.7-mm diameter lens system was designed and combined with the MEMS mirror to achieve a small form factor that optimize functionality as a handheld extravascular OCT imager. To evaluate in-vivo applicability, super-microsurgical vessel anastomosis was performed in a mouse femoral vessel cut and repair model employing conventional interrupted suture technique as well as a novel non-suture cuff technique. Vascular anastomosis patency after clinically successful repair was evaluated using the novel handheld OCT imager.ResultsWith an adjustable lateral image field of view up to 1.5 mm by 1.5 mm, high-resolution simultaneous structural and flow imaging of the blood vessels were successfully acquired for BALB/C mouse after orthotopic hind limb transplantation using a non-suture cuff technique and BALB/C mouse after femoral artery anastomosis using a suture technique. We experimentally quantify the axial and lateral resolution of the OCT to be 12.6 µm in air and 17.5 µm respectively. The OCT has a sensitivity of 84 dB and sensitivity roll-off of 5.7 dB/mm over an imaging range of 5 mm. Imaging with a frame rate of 36 Hz for an image size of 1000(lateral)×512(axial) pixels using a 50,000 A-lines per second swept source was achieved. Quantitative vessel lumen patency, lumen narrowing and thrombosis analysis were performed based on acquired structure and Doppler images.ConclusionsA miniature handheld OCT imager that can be used for intraoperative evaluation of microvascular anastomosis was successfully demonstrated.

Dataset Information

Handheld Briefcase Optical Coherence Tomography with Real-Time Machine Learning Classifier for Middle Ear Infections.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets