Modification of a Conventional Deep Learning Model to Classify Simulated Breathing Patterns: A Step toward Real-Time Monitoring of Patients with Respiratory Infectious Diseases.

The emergence of the global coronavirus pandemic in 2019 (COVID-19 disease) created a need for remote methods to detect and continuously monitor patients with infectious respiratory diseases. Many different devices, including thermometers, pulse oximeters, smartwatches, and rings, were proposed to monitor the symptoms of infected individuals at home. However, these consumer-grade devices are typically not capable of automated monitoring during both day and night. This study aims to develop a method to classify and monitor breathing patterns in real-time using tissue hemodynamic responses and a deep convolutional neural network (CNN)-based classification algorithm. Tissue hemodynamic responses at the sternal manubrium were collected in 21 healthy volunteers using a wearable near-infrared spectroscopy (NIRS) device during three different breathing conditions. We developed a deep CNN-based classification algorithm to classify and monitor breathing patterns in real time. The classification method was designed by improving and modifying the pre-activation residual network (Pre-ResNet) previously developed to classify two-dimensional (2D) images. Three different one-dimensional CNN (1D-CNN) classification models based on Pre-ResNet were developed. By using these models, we were able to obtain an average classification accuracy of 88.79% (without Stage 1 (data size reducing convolutional layer)), 90.58% (with 1 × 3 Stage 1), and 91.77% (with 1 × 5 Stage 1).

Optical Monitoring of Breathing Patterns and Tissue Oxygenation: A Potential Application in COVID-19 Screening and Monitoring.

The worldwide outbreak of the novel Coronavirus (COVID-19) has highlighted the need for a screening and monitoring system for infectious respiratory diseases in the acute and chronic phase. The purpose of this study was to examine the feasibility of using a wearable near-infrared spectroscopy (NIRS) sensor to collect respiratory signals and distinguish between normal and simulated pathological breathing. Twenty-one healthy adults participated in an experiment that examined five separate breathing conditions. Respiratory signals were collected with a continuous-wave NIRS sensor (PortaLite, Artinis Medical Systems) affixed over the sternal manubrium. Following a three-minute baseline, participants began five minutes of imposed difficult breathing using a respiratory trainer. After a five minute recovery period, participants began five minutes of imposed rapid and shallow breathing. The study concluded with five additional minutes of regular breathing. NIRS signals were analyzed using a machine learning model to distinguish between normal and simulated pathological breathing. Three features: breathing interval, breathing depth, and O2Hb signal amplitude were extracted from the NIRS data and, when used together, resulted in a weighted average accuracy of 0.87. This study demonstrated that a wearable NIRS sensor can monitor respiratory patterns continuously and non-invasively and we identified three respiratory features that can distinguish between normal and simulated pathological breathing.

Evaluation of neurocognitive function of prefrontal cortex in ornithine transcarbamylase deficiency.

Hyperammonia due to ornithine transcarbamylase deficiency (OTCD) can cause a range of deficiencies in domains of executive function and working memory. Only a few fMRI studies have focused on neuroimaging data in a population with OTCD. Yet, there is a need for monitoring the disease progression and neurocognitive function in this population. In this study, we used a non-invasive neuroimaging technique, functional Near Infrared Spectroscopy (fNIRS), to examine the hemodynamics of prefrontal cortex (PFC) based on neural activation in an OTCD population. Using fNIRS, we measured the activation in PFC of the participants while performing the Stroop task. Behavioral assessment such as reaction time and correct response were recorded. We investigated the difference in behavioral measures as well as brain activation in left and right PFC in patients with OTCD and controls. Results revealed a distinction in left PFC activation between controls and patients with OTCD, where control subjects showed higher task related activation increase. Subjects with OTCD also exhibited bilateral increase in PFC activation. There was no significant difference in response time or correct response between the two groups. Our findings suggest the alterations in neurocognitive function of PFC in OTCD compared to the controls despite the behavioral profiles exhibiting no such differences. This is a first study using fNIRS to examine a neurocognitive function in OTCD population and can provide a novel insight into the screening of OTCD progression and examining neurocognitive changes.

Functional near-infrared spectroscopy in toddlers: Neural differentiation of communicative cues and relation to future language abilities.

The toddler and preschool years are a time of significant development in both expressive and receptive communication abilities. However, little is known about the neurobiological underpinnings of language development during this period, likely due to difficulties acquiring functional neuroimaging data. Functional near-infrared spectroscopy (fNIRS) is a motion-tolerant neuroimaging technique that assesses cortical brain activity and can be used in very young children. Here, we use fNIRS during perception of communicative and noncommunicative speech and gestures in typically developing 2- and 3-year-olds (Study 1, n = 15, n = 12 respectively) and in a sample of 2-year-olds with both fNIRS data collected at age 2 and language outcome data at age 3 (Study 2, n = 18). In Study 1, 2- and 3-year-olds differentiated between communicative and noncommunicative stimuli as well as between speech and gestures in the left lateral frontal region. However, 2-year-olds showed different patterns of activation from 3-year-olds in right medial frontal regions. In Study 2, which included two toddlers identified with early language delays along with 16 typically developing toddlers, neural differentiation of communicative stimuli in the right medial frontal region at age 2 predicted receptive language at age 3. Specifically, after accounting for variance related to verbal ability at age 2, increased neural activation for communicative gestures (vs. both communicative speech and noncommunicative gestures) at age 2 predicted higher receptive language scores at age 3. These results are discussed in the context of the underlying mechanisms of toddler language development and use of fNIRS in prediction of language outcomes.

Optical Imaging Technology: A Useful Tool to Identify Remission in Cushing Disease After Surgery.

We recently reported the use of optical imaging technology to quantify facial plethora in endogenous Cushing syndrome (CS). In the present study, we studied a larger cohort of patients with Cushing disease (CD) and examined water content fraction as well as blood volume fraction as bio-optic markers for determining the efficacy of this methodology as a predictor of lasting remission after surgery for CS. We imaged 49 patients before and after transsphenoidal surgery (TSS) for Cushing disease (CD); 22 patients were also seen at 3-6 months, and 13 patients 12 months post-operatively. On all patients, we used multi-spectral imaging (MSI) to evaluate hemodynamic distributions as well as water content at a specific area of the face. We found a decrease in blood volume fraction after vs. before surgical treatment in the tested facial area in 37 of the 40 patients, as determined with biochemical markers (p<0.001). All patients that were followed up for up to 12 months showed the same decrease from preoperative values and they remained in remission from CD. We conclude that MSI can be used for the evaluation of remission from CD, at least in the immediate post-operative period and up to one year after surgery. The use of this technology can supplement biochemical and other testing for the evaluation of the various treatment modalities available for patients with CD.

Multiscale BerEp4 molecular imaging of microtumor phantoms: toward theranostics for basal cell carcinoma.

Basal cell carcinoma (BCC), the most common cancer in humans, appears macroscopically and microscopically similar to many other skin lesions, which makes differential diagnosis difficult. We are developing an approach for quantitative molecular imaging of BerEP4, a transmembrane biomarker for BCC, with the goal of increasing the precision and accuracy of diagnosis. This pilot study was conducted to assess the affinity and selectivity of BerEp4 antibody and assess its possible use in designing theranostic probes for BCC. We provide evidence that our photon-counting fluorescence macrodetection system can recover specific signal increases from a film/pellet phantom. Additionally, we show that a two-photon excited fluorescence /backscatter confocal microscopy system can image BerEP4 antibody/antigen complex on the surface of BerEP4-expressing cancer cells in three dimensions. Based on the initial results, BerEP4 seems to be a promising biomarker for molecular imaging of BCC. To prepare BerEP4 for eventual theranostic use, we examined the feasibility of a combined macro-/micro-optical approach to imaging BCC with various histologies. These optical methods, endowed with the ability to monitor treatment in real time, may open an opportunity for noninvasive diagnosis, treatments, and follow-up.

Compact non-contact total emission detection for in vivo multiphoton excitation microscopy.

We describe a compact, non-contact design for a total emission detection (c-TED) system for intra-vital multiphoton imaging. To conform to a standard upright two-photon microscope design, this system uses a parabolic mirror surrounding a standard microscope objective in concert with an optical path that does not interfere with normal microscope operation. The non-contact design of this device allows for maximal light collection without disrupting the physiology of the specimen being examined. Tests were conducted on exposed tissues in live animals to examine the emission collection enhancement of the c-TED device compared to heavily optimized objective-based emission collection. The best light collection enhancement was seen from murine fat (5×-2× gains as a function of depth), whereas murine skeletal muscle and rat kidney showed gains of over two and just under twofold near the surface, respectively. Gains decreased with imaging depth (particularly in the kidney). Zebrafish imaging on a reflective substrate showed close to a twofold gain throughout the entire volume of an intact embryo (approximately 150 µm deep). Direct measurement of bleaching rates confirmed that the lower laser powers, enabled by greater light collection efficiency, yielded reduced photobleaching in vivo. The potential benefits of increased light collection in terms of speed of imaging and reduced photo-damage, as well as the applicability of this device to other multiphoton imaging methods is discussed.

Application of the Single Source-Detector Separation Algorithm in Wearable Neuroimaging Devices: A Step toward Miniaturized Biosensor for Hypoxia Detection.

Most currently available wearable devices to noninvasively detect hypoxia use the spatially resolved spectroscopy (SRS) method to calculate cerebral tissue oxygen saturation (StO2). This study applies the single source-detector separation (SSDS) algorithm to calculate StO2. Near-infrared spectroscopy (NIRS) data were collected from 26 healthy adult volunteers during a breath-holding task using a wearable NIRS device, which included two source-detector separations (SDSs). These data were used to derive oxyhemoglobin (HbO) change and StO2. In the group analysis, both HbO change and StO2 exhibited significant change during a breath-holding task. Specifically, they initially decreased to minimums at around 10 s and then steadily increased to maximums, which were significantly greater than baseline levels, at 25-30 s (p-HbO < 0.001 and p-StO2 < 0.05). However, at an individual level, the SRS method failed to detect changes in cerebral StO2 in response to a short breath-holding task. Furthermore, the SSDS algorithm is more robust than the SRS method in quantifying change in cerebral StO2 in response to a breath-holding task. In conclusion, these findings have demonstrated the potential use of the SSDS algorithm in developing a miniaturized wearable biosensor to monitor cerebral StO2 and detect cerebral hypoxia.

Developmental Differences in Reaching-and-Placing Movement and Its Potential in Classifying Children with and without Autism Spectrum Disorder: Deep Learning Approach.

Autism Spectrum Disorder (ASD) is among the most prevalent neurodevelopmental disorders, yet the current diagnostic procedures rely on behavioral analyses and interviews and lack objective screening methods. This study seeks to address this gap by integrating upper limb kinematics and deep learning methods to identify potential biomarkers that could be validated in younger age groups in the future to enhance the identification of ASD. Forty-one school-age children, with and without an ASD diagnosis (Mean age ± SE = 10.3 ± 0.4; 12 Females), participated in the study. A single Inertial Measurement Unit (IMU) was affixed to the child's wrist as they engaged in a continuous reaching and placing task. Deep learning techniques were employed to classify children with and without ASD. Our findings suggest delays in motor planning and control in school-age children compared to healthy adults. Compared to TD children, children with ASD exhibited poor motor planning and control as seen by greater number of movement units, more movement overshooting, and prolonged time to peak velocity/acceleration. Compensatory movement strategies such as greater velocity and acceleration were also seen in the ASD group. More importantly, using Multilayer Perceptron (MLP) model, we demonstrated an accuracy of ~ 78.1% in classifying children with and without ASD. These findings underscore the potential use of studying upper limb movement kinematics during goal-directed arm movements and deep learning methods as valuable tools for classifying and, consequently, aiding in the diagnosis and early identification of ASD upon further validation in younger children.

Visual stimulus structure, visual system neural activity, and visual behavior in young human infants.

In visual perception and information processing, a cascade of associations is hypothesized to flow from the structure of the visual stimulus to neural activity along the retinogeniculostriate visual system to behavior and action. Do visual perception and information processing adhere to this cascade near the beginning of life? To date, this three-stage hypothetical cascade has not been comprehensively tested in infants. In two related experiments, we attempted to expose this cascade in 6-month-old infants. Specifically, we presented infants with two levels of visual stimulus intensity, we measured electrical activity at the infant cortex, and we assessed infants' preferential looking behavior. Chromatic saturation provided a convenient stimulus dimension to test the cascade because greater saturation is known to excite increased activity in the primate visual system and is generally hypothesized to stimulate visual preference. Experiment 1 revealed that infants prefer (look longer) at the more saturated of two colors otherwise matched in hue and brightness. Experiment 2 showed increased aggregate neural cortical excitation in infants (and adults) to the more saturated of the same pair of colors. Thus, experiments 1 and 2 taken together confirm a cascade: Visual stimulation of relatively greater intensity evokes relatively greater levels of bioelectrical cortical activity which in turn is associated with relatively greater visual attention. As this cascade obtains near the beginning of life, it helps to account for early visual preferences and visual information processing.

Capturing dynamic patterns of task-based functional connectivity with EEG.

A new approach to trace the dynamic patterns of task-based functional connectivity, by combining signal segmentation, dynamic time warping (DTW), and Quality Threshold (QT) clustering techniques, is presented. Electroencephalography (EEG) signals of 5 healthy subjects were recorded as they performed an auditory oddball and a visual modified oddball tasks. To capture the dynamic patterns of functional connectivity during the execution of each task, EEG signals are segmented into durations that correspond to the temporal windows of previously well-studied event-related potentials (ERPs). For each temporal window, DTW is employed to measure the functional similarities among channels. Unlike commonly used temporal similarity measures, such as cross correlation, DTW compares time series by taking into consideration that their alignment properties may vary in time. QT clustering analysis is then used to automatically identify the functionally connected regions in each temporal window. For each task, the proposed approach was able to establish a unique sequence of dynamic pattern (observed in all 5 subjects) for brain functional connectivity.

Electrophysiological Evidence for Interhemispheric Connectivity and Communication in Young Human Infants.

Little is known empirically about connectivity and communication between the two hemispheres of the brain in the first year of life, and what theoretical opinion exists appears to be at variance with the meager extant anatomical evidence. To shed initial light on the question of interhemispheric connectivity and communication, this study investigated brain correlates of interhemispheric transmission of information in young human infants. We analyzed EEG data from 12 4-month-olds undergoing a face-related oddball ERP protocol. The activity in the contralateral hemisphere differed between odd-same and odd-difference trials, with the odd-different response being weaker than the response during odd-same trials. The infants' contralateral hemisphere "recognized" the odd familiar stimulus and "discriminated" the odd-different one. These findings demonstrate connectivity and communication between the two hemispheres of the brain in the first year of life and lead to a better understanding of the functional integrity of the developing human infant brain.

The use of functional near-infrared spectroscopy in tracking neurodevelopmental trajectories in infants and children with or without developmental disorders: a systematic review.

Understanding the neurodevelopmental trajectories of infants and children is essential for the early identification of neurodevelopmental disorders, elucidating the neural mechanisms underlying the disorders, and predicting developmental outcomes. Functional Near-Infrared Spectroscopy (fNIRS) is an infant-friendly neuroimaging tool that enables the monitoring of cerebral hemodynamic responses from the neonatal period. Due to its advantages, fNIRS is a promising tool for studying neurodevelopmental trajectories. Although many researchers have used fNIRS to study neural development in infants/children and have reported important findings, there is a lack of synthesized evidence for using fNIRS to track neurodevelopmental trajectories in infants and children. The current systematic review summarized 84 original fNIRS studies and showed a general trend of age-related increase in network integration and segregation, interhemispheric connectivity, leftward asymmetry, and differences in phase oscillation during resting-state. Moreover, typically developing infants and children showed a developmental trend of more localized and differentiated activation when processing visual, auditory, and tactile information, suggesting more mature and specialized sensory networks. Later in life, children switched from recruiting bilateral auditory to a left-lateralized language circuit when processing social auditory and language information and showed increased prefrontal activation during executive functioning tasks. The developmental trajectories are different in children with developmental disorders, with infants at risk for autism spectrum disorder showing initial overconnectivity followed by underconnectivity during resting-state; and children with attention-deficit/hyperactivity disorders showing lower prefrontal cortex activation during executive functioning tasks compared to their typically developing peers throughout childhood. The current systematic review supports the use of fNIRS in tracking the neurodevelopmental trajectories in children. More longitudinal studies are needed to validate the neurodevelopmental trajectories and explore the use of these neurobiomarkers for the early identification of developmental disorders and in tracking the effects of interventions.

Skin Lesion Correspondence Localization in Total Body Photography.

Longitudinal tracking of skin lesions - finding correspondence, changes in morphology, and texture - is beneficial to the early detection of melanoma. However, it has not been well investigated in the context of full-body imaging. We propose a novel framework combining geometric and texture information to localize skin lesion correspondence from a source scan to a target scan in total body photography (TBP). Body landmarks or sparse correspondence are first created on the source and target 3D textured meshes. Every vertex on each of the meshes is then mapped to a feature vector characterizing the geodesic distances to the landmarks on that mesh. Then, for each lesion of interest (LOI) on the source, its corresponding location on the target is first coarsely estimated using the geometric information encoded in the feature vectors and then refined using the texture information. We evaluated the framework quantitatively on both a public and a private dataset, for which our success rates (at 10 mm criterion) are comparable to the only reported longitudinal study. As full-body 3D capture becomes more prevalent and has higher quality, we expect the proposed method to constitute a valuable step in the longitudinal tracking of skin lesions.

Simultaneous multimodal fNIRS-EEG recordings reveal new insights in neural activity during motor execution, observation, and imagery.

Motor execution, observation, and imagery are important skills used in motor learning and rehabilitation. The neural mechanisms underlying these cognitive-motor processes are still poorly understood. We used a simultaneous recording of functional near-infrared spectroscopy (fNIRS) and electroencephalogram (EEG) to elucidate the differences in neural activity across three conditions requiring these processes. Additionally, we used a new method called structured sparse multiset Canonical Correlation Analysis (ssmCCA) to fuse the fNIRS and EEG data and determine the brain regions of neural activity consistently detected by both modalities. Unimodal analyses revealed differentiated activation between conditions; however, the activated regions did not fully overlap across the two modalities (fNIRS: left angular gyrus, right supramarginal gyrus, as well as right superior and inferior parietal lobes; EEG: bilateral central, right frontal, and parietal). These discrepancies might be because fNIRS and EEG detect different signals. Using fused fNIRS-EEG data, we consistently found activation over the left inferior parietal lobe, superior marginal gyrus, and post-central gyrus during all three conditions, suggesting that our multimodal approach identifies a shared neural region associated with the Action Observation Network (AON). This study highlights the strengths of using the multimodal fNIRS-EEG fusion technique for studying AON. Neural researchers should consider using the multimodal approach to validate their findings.

Normative database of judgment of complexity task with functional near infrared spectroscopy--application for TBI.

The ability to assess frontal lobe function in a rapid, objective, and standardized way, without the need for expertise in cognitive test administration might be particularly helpful in mild traumatic brain injury (TBI), where objective measures are needed. Functional near infrared spectroscopy (fNIRS) is a reliable technique to noninvasively measure local hemodynamic changes in brain areas near the head surface. In this paper, we are combining fNIRS and frameless stereotaxy which allowed us to co-register the functional images with previously acquired anatomical MRI volumes. In our experiment, the subjects were asked to perform a task, evaluating the complexity of daily life activities, previously shown with fMRI to activate areas of the anterior frontal cortex. We reconstructed averaged oxyhemoglobin and deoxyhemoglobin data from 20 healthy subjects in a spherical coordinate. The spherical coordinate is a natural representation of surface brain activation projection. Our results show surface activation projected from the medial frontopolar cortex which is consistent with previous fMRI results. With this original technique, we will construct a normative database for a simple cognitive test which can be useful in evaluating cognitive disability such as mild traumatic brain injury.

In vivo method to monitor changes in HER2 expression using near-infrared fluorescence imaging.

Human epidermal growth factor receptor type 2 (HER2) is a well-known biomarker that is overexpressed in many breast carcinomas. HER2 expression level is an important factor to optimize the therapeutic strategy and monitor the treatment. We used albumin binding domain-fused HER2-specific Affibody molecules, labeled with Alexa Fluor750 dye, to characterize HER2 expression in vivo. Near-infrared optical imaging studies were carried out using mice with subcutaneous HER2-positive tumors. Animals were divided into groups of five: no treatment and 12 hours and 1 week after treatment of the tumors with the Hsp90 inhibitor 17-dimethylaminoethylamino-17-demethoxygeldanamycin (17-DMAG). The compartmental ligands-receptor model, describing binding kinetics, was used to evaluate HER2 expression from the time sequence of the fluorescence images after the intravenous probe injection. The normalized rate of accumulation of the specific fluorescent biomarkers, estimated from this time sequence, linearly correlates with the conventional ex vivo enzyme-linked immunosorbent assay (ELISA) readings for the same tumor. Such correspondence makes properly arranged fluorescence imaging an excellent candidate for estimating HER2 overexpression in tumors, complementing ELISA and other ex vivo assays. Application of this method to the fluorescence data from HER2-positive xenografts reveals that the 17-DMAG treatment results in downregulation of HER2. Application of the AngioSense 750 probe confirmed the antiangiogenic effect of 17-DMAG found with Affibody-Alexa Fluor 750 conjugate.

Label-free, non-invasive, and repeatable cell viability bioassay using dynamic full-field optical coherence microscopy and supervised machine learning.

We present a novel method that can assay cellular viability in real-time using supervised machine learning and intracellular dynamic activity data that is acquired in a label-free, non-invasive, and non-destructive manner. Cell viability can be an indicator for cytology, treatment, and diagnosis of diseases. We applied four supervised machine learning models on the observed data and compared the results with a trypan blue assay. The cell death assay performance by the four supervised models had a balanced accuracy of 93.92 ± 0.86%. Unlike staining techniques, where criteria for determining viability of cells is unclear, cell viability assessment using machine learning could be clearly quantified.

Cerebrovascular Reactivity Measurement with Functional Near Infrared Spectroscopy.

Cerebrovascular reactivity (CVR) is the capacity of blood vessels in the brain to alter cerebral blood flow (either with dilation or constriction) in response to chemical or physical stimuli. The amount of reactivity in the cerebral microvasculature depends on the integrity of the capacitance vasculature and is the primary function of endothelial cells. CVR is, therefore, an indicator of the microvasculature's physiology and overall health. Imaging methods that can measure CVR are available but can be costly, and require magnetic resonance imaging centers and technical expertise. In this study, we used fNIRS technology to monitor changes of oxyhemoglobin (HbO) and deoxyhemoglobin (HbR) in the cerebral microvasculature to assess the CVR of 15 healthy controls (HC) in response to a vasoactive stimulus (inhaled 5% carbon dioxide or CO2). Our results suggest that this is a promising imaging technology that offers a non-invasive, accurate, portable, and cost-effective method of mapping cortical CVR and associated microvasculature function, resulting from a traumatic brain injury or other conditions associated with cerebral microvasculopathy.