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.

No abstract

Polarization-sensitive optical coherence tomography (PS-OCT) is a high-resolution label-free optical biomedical imaging modality that is sensitive to the microstructural architecture in tissue that gives rise to form birefringence, such as collagen or muscle fibers. To enable polarization sensitivity in an OCT system, however, requires additional hardware and complexity. We developed a deep-learning method to synthesize PS-OCT images by training a generative adversarial network (GAN) on OCT intensity and PS-OCT images. The synthesis accuracy was first evaluated by the structural similarity index (SSIM) between the synthetic and real PS-OCT images. Furthermore, the effectiveness of the computational PS-OCT images was validated by separately training two image classifiers using the real and synthetic PS-OCT images for cancer/normal classification. The similar classification results of the two trained classifiers demonstrate that the predicted PS-OCT images can be potentially used interchangeably in cancer diagnosis applications. In addition, we applied the trained GAN models on OCT images collected from a separate OCT imaging system, and the synthetic PS-OCT images correlate well with the real PS-OCT image collected from the same sample sites using the PS-OCT imaging system. This computational PS-OCT imaging method has the potential to reduce the cost, complexity, and need for hardware-based PS-OCT imaging systems.

Understanding drug fingerprints in complex biological samples is essential for the development of a drug. Hyperspectral coherent anti-Stokes Raman scattering (HS-CARS) microscopy, a label-free nondestructive chemical imaging technique, can profile biological samples based on their endogenous vibrational contrast. Here, we propose a deep learning-assisted HS-CARS imaging approach for the investigation of drug fingerprints and their localization at single-cell resolution. To identify and localize drug fingerprints in complex biological systems, an attentionbased deep neural network, hyperspectral attention net (HAN), was developed. By formulating the task to a multiple instance learning problem, HAN highlights informative regions through the attention mechanism when being trained on wholeimage labels. Using the proposed technique, we investigated the drug fingerprints of a hepatitis B virus therapy in murine liver tissues. With the increase in drug dosage, higher classification accuracy was observed, with an average area under the curve (AUC) of 0.942 for the high-dose group. Besides, highly informative tissue structures predicted by HAN demonstrated a high degree of similarity with the drug localization shown by the in situ hybridization staining results. These results demonstrate the potential of the proposed deep learning-assisted optical imaging technique for the label-free profiling, identification, and localization of drug fingerprints in biological samples, which can be extended to nonperturbative investigations of complex biological systems under various biological conditions.

With the latest advancements in optical bioimaging, rich structural and functional information is generated from biological samples, which calls for powerful computational tools to identify patterns and uncover relationships between optical characteristics and various biological conditions. Here we present a weakly supervised deep learning framework for human breast cancer-related optical signature discovery based on virtual histopathology enabled by simultaneous label-free autofluorescence multiharmonic microscopy. This framework consists of (1) cancer diagnosis by leveraging inexact and incomplete supervision, and (2) cancer-related optical signature identification via the integration of multiple model interpretation techniques. The framework has achieved an average area under the curve (AUC) of 0.975 on the cancer diagnosis task. In addition to well-known cancer biomarkers, non-obvious cancer-related patterns were revealed by the framework, including NAD(P)H-rich extracellular vesicles observed in normal-appearing breast cancer tissue, which facilitate new insights into the tumor microenvironment and field cancerization. This framework can be further extended to diverse optical signature discovery tasks and a variety of imaging modalities.