Repeatability and reproducibility of quantitative OCT angiography measurements from table-top and portable Flex Spectralis devices.

PURPOSE: The recent development of a portable investigational handheld OCT-angiography (OCTA) device has allowed for expansion of imaging into the operating room (OR) in addition to standard in-clinic imaging. The aim of this study was to assess intravisit repeatability and intervisit reproducibility of retinal microvasculature measures and central retinal thickness for in-clinic table-top and portable OR compatible OCTA devices. METHODS: Repeated 10 × 10 OCTA images were acquired in 20 healthy adult participants on two separate visit days using Spectralis spectral-domain OCTA table-top and investigational armature suspended Flex systems. Intravisit and intervisit intraclass correlation coefficients and average absolute percent difference were calculated for quantitative microvasculature measures and CRT. RESULTS: 120 OCTA images were acquired from 20 subjects (n = 20, mean age 26.7 ± 1.61 years, range 24-30 years) with both devices across two separate imaging days. FAZ and CRT measurements had near complete intravisit and intervisit agreement with ICCs between .97 and 1 for both table-top (FAZ ICC .97, .97; CRT ICC .98-1, .98-.99) and Flex (FAZ ICC .97, .95; CRT ICC .99-1, .98-.99) devices. Vessel density measures demonstrated greater variance with only fair to strong agreement (ICC .32-.75) and average absolute percent differences ranging from 2.96 to 6.63%. CONCLUSION: FAZ and CRT measures for both devices demonstrated high repeatability and reproducibility; retinal vessel density measures demonstrated less. Differences of less than 7% for retinal microvasculature measurements across time and devices are most likely attributable to expectable variance between repeat scans.

Sensitive and Quantitative Point-of-Care HIV Viral Load Quantification from Blood Using a Power-Free Plasma Separation and Portable Magnetofluidic Polymerase Chain Reaction Instrument.

Point-of-care (POC) HIV viral load (VL) tests are needed to enhance access to HIV VL testing in low- and middle-income countries (LMICs) and to enable HIV VL self-testing at home, which in turn have the potential to enhance the global management of the disease. While methods based on real-time reverse transcription-polymerase chain reaction (RT-PCR) are highly sensitive and quantitatively accurate, they often require bulky and expensive instruments, making applications at the POC challenging. On the other hand, although methods based on isothermal amplification techniques could be performed using low-cost instruments, they have shown limited quantitative accuracies, i.e., being only semiquantitative. Herein, we present a sensitive and quantitative POC HIV VL quantification method from blood that can be performed using a small power-free three-dimensional-printed plasma separation device and a portable, low-cost magnetofluidic real-time RT-PCR instrument. The plasma separation device, which is composed of a plasma separation membrane and an absorbent material, demonstrated 96% plasma separation efficiency per 100 µL of whole blood. The plasma solution was then processed in a magnetofluidic cartridge for automated HIV RNA extraction and quantification using the portable instrument, which completed 50 cycles of PCR in 15 min. Using the method, we achieved a limit of detection of 500 HIV RNA copies/mL, which is below the World Health Organization's virological failure threshold, and a good quantitative accuracy. The method has the potential for sensitive and quantitative HIV VL testing at the POC and at home self-testing.

Listeria monocytogenes exploits host exocytosis to promote cell-to-cell spread.

The facultative intracellular pathogen Listeria monocytogenes uses an actin-based motility process to spread within human tissues. Filamentous actin from the human cell forms a tail behind bacteria, propelling microbes through the cytoplasm. Motile bacteria remodel the host plasma membrane into protrusions that are internalized by neighboring cells. A critical unresolved question is whether generation of protrusions by Listeria involves stimulation of host processes apart from actin polymerization. Here we demonstrate that efficient protrusion formation in polarized epithelial cells involves bacterial subversion of host exocytosis. Confocal microscopy imaging indicated that exocytosis is up-regulated in protrusions of Listeria in a manner that depends on the host exocyst complex. Depletion of components of the exocyst complex by RNA interference inhibited the formation of Listeria protrusions and subsequent cell-to-cell spread of bacteria. Additional genetic studies indicated important roles for the exocyst regulators Rab8 and Rab11 in bacterial protrusion formation and spread. The secreted Listeria virulence factor InlC associated with the exocyst component Exo70 and mediated the recruitment of Exo70 to bacterial protrusions. Depletion of exocyst proteins reduced the length of Listeria protrusions, suggesting that the exocyst complex promotes protrusion elongation. Collectively, these results demonstrate that Listeria exploits host exocytosis to stimulate intercellular spread of bacteria.

Role of septins in microbial infection.

Septins are widely recognized as a component of the cytoskeleton that is essential for cell division, and new work has shown that septins can recognise cell shape by assembling into filaments on membrane regions that display micrometer-scale curvature (e.g. at the cytokinetic furrow). Moreover, infection biology studies have illuminated important roles for septins in mediating the outcome of host-microbe interactions. In this Review, we discuss a selection of mechanistic insights recently gained from studying three infection paradigms: the rice blast fungus Magnaporthe oryzae, the poxvirus family member vaccinia virus and the Gram-negative bacterium Shigella flexneri These studies have respectively discovered that higher-order septin assemblies enable fungal invasion into plant cells, entrap viral particles at the plasma membrane and recognize dividing bacterial cells for delivery to lysosomes. Collectively, these insights illustrate how studying septin biology during microbial infection can provide fundamental advances in both cell and infection biology, and suggest new concepts underlying infection control.

The Host Scaffolding Protein Filamin A and the Exocyst Complex Control Exocytosis during InlB-Mediated Entry of Listeria monocytogenes.

Listeria monocytogenes is a foodborne bacterium that causes gastroenteritis, meningitis, or abortion. Listeria induces its internalization (entry) into some human cells through interaction of the bacterial surface protein InlB with its host receptor, the Met tyrosine kinase. InlB and Met promote entry, in part, through stimulation of localized exocytosis. How exocytosis is upregulated during entry is not understood. Here, we show that the human signaling proteins mTOR, protein kinase C-α (PKC-α), and RalA promote exocytosis during entry by controlling the scaffolding protein Filamin A (FlnA). InlB-mediated uptake was accompanied by PKC-α-dependent phosphorylation of serine 2152 in FlnA. Depletion of FlnA by RNA interference (RNAi) or expression of a mutated FlnA protein defective in phosphorylation impaired InlB-dependent internalization. These findings indicate that phosphorylation of FlnA by PKC-α contributes to entry. mTOR and RalA were found to mediate the recruitment of FlnA to sites of InlB-mediated entry. Depletion of PKC-α, mTOR, or FlnA each reduced exocytosis during InlB-mediated uptake. Because the exocyst complex is known to mediate polarized exocytosis, we examined if PKC-α, mTOR, RalA, or FlnA affects this complex. Depletion of PKC-α, mTOR, RalA, or FlnA impaired recruitment of the exocyst component Exo70 to sites of InlB-mediated entry. Experiments involving knockdown of Exo70 or other exocyst proteins demonstrated an important role for the exocyst complex in uptake of Listeria Collectively, our results indicate that PKC-α, mTOR, RalA, and FlnA comprise a signaling pathway that mobilizes the exocyst complex to promote infection by Listeria.

Assessment of Macular Microvasculature in Healthy Eyes of Infants and Children Using OCT Angiography.

PURPOSE: To assess macular vasculature in healthy infants and children using OCT angiography (OCTA). DESIGN: Prospective cross-sectional study. PARTICIPANTS: One hundred thirty-five normal maculae of 89 healthy infants and children (mean age, 8.5±5.3 years; range, 9 weeks-17 years) treated at the Duke University Eye Center. METHODS: We imaged 135 maculae of 89 pediatric patients using the standard Spectralis tabletop and investigational Spectralis with Flex module devices, both equipped with investigational OCTA software (Heidelberg Engineering, Heidelberg, Germany). OCT angiography images of the superficial vascular complex (SVC) and deep vascular complex (DVC) were analyzed for foveal avascular zone (FAZ) area and superficial and deep vessel density. We assessed effects of age, gender, race, axial length (AL), and central subfield thickness on FAZ and vessel density. Patients with both eyes imaged were assessed for agreement between the FAZ and vessel densities of the left and right eyes. MAIN OUTCOME MEASURES: The FAZ area, as well as vessel area density (VAD) and vessel length density (VLD) in the SVC and DVC. RESULTS: The FAZ varied significantly with race; white patients showed a significantly smaller FAZ than black patients (mean difference, 0.11 mm2; P = 0.004). The FAZ did not vary with age, gender, or AL (P > 0.05). In the SVC, VAD and VLD varied significantly with age (P < 0.001) and AL (R2 = 0.46; P < 0.001) but not gender (P > 0.05). The SVC VLD was significantly different between races and ethnicities (P = 0.037), but VAD was not (P < 0.05). In the DVC, VAD and VLD also varied significantly with age (P < 0.001) and AL (R2 = 0.46; P < 0.001) but not gender or race (P > 0.05). There was excellent agreement between the right and left eyes for FAZ (intraclass correlation [ICC], 0.97), SVC VLD (ICC, 1.00), and DVC VLD (ICC, 1.00). CONCLUSIONS: Quantitative studies of pediatric perifoveal vasculature should consider age, race, and AL. In eyes with unilateral disease, the perifoveal vasculature in the unaffected eye may be used as a control comparison because there is excellent agreement between eyes.

Interaction of microbial pathogens with host exocytic pathways.

Many microbial pathogens co-opt or perturb host membrane trafficking pathways. This review covers recent examples in which microbes interact with host exocytosis, the fusion of intracellular vesicles with the plasma membrane. The bacterial pathogens Listeria monocytogenes and Staphylococcus aureus subvert recycling endosomal pathways of exocytosis in order to induce their entry into human cells. By contrast, entry of the protozoan pathogen Trypanosoma cruzi or the virus adenovirus into host cells involves exploitation of lysosomal exocytosis. Toxins produced by Bacillus anthracis or Vibrio cholerae interfere with exocytosis pathways mediated by the GTPase Rab11 and the exocyst complex. By doing so, anthrax or cholera toxins impair recycling of cadherins to cell-cell junctions and disrupt the barrier properties of endothelial cells or intestinal epithelial cells, respectively. Uropathogenic Escherichia coli (UPEC) is expelled from bladder epithelial cells through two different exocytic routes that involve sensing of bacteria in vacuoles by host Toll-like receptor 4 (TLR4) or monitoring of the pH of lysosomes harbouring UPEC. The TLR4 pathway is mediated by multiple Rab GTPases and the exocyst, whereas the other pathway involves exocytosis of lysosomes. Expulsion of UPEC through these pathways is thought to benefit the host.

A role for host cell exocytosis in InlB-mediated internalisation of Listeria monocytogenes.

The bacterial surface protein InlB mediates internalisation of Listeria monocytogenes into human cells through interaction with the host receptor tyrosine kinase, Met. InlB-mediated entry requires localised polymerisation of the host actin cytoskeleton. Apart from actin polymerisation, roles for other host processes in Listeria entry are unknown. Here, we demonstrate that exocytosis in the human cell promotes InlB-dependent internalisation. Using a probe consisting of VAMP3 with an exofacial green fluorescent protein tag, focal exocytosis was detected during InlB-mediated entry. Exocytosis was dependent on Met tyrosine kinase activity and the GTPase RalA. Depletion of SNARE proteins by small interfering RNA demonstrated an important role for exocytosis in Listeria internalisation. Depletion of SNARE proteins failed to affect actin filaments during internalisation, suggesting that actin polymerisation and exocytosis are separable host responses. SNARE proteins were required for delivery of the human GTPase Dynamin 2, which promotes InlB-mediated entry. Our results identify exocytosis as a novel host process exploited by Listeria for infection.

Plasmonic SERS biosensing nanochips for DNA detection.

The development of rapid, cost-effective DNA detection methods for molecular diagnostics at the point-of-care (POC) has been receiving increasing interest. This article reviews several DNA detection techniques based on plasmonic-active nanochip platforms developed in our laboratory over the last 5 years, including the molecular sentinel-on-chip (MSC), the multiplex MSC, and the inverse molecular sentinel-on-chip (iMS-on-Chip). DNA probes were used as the recognition elements, and surface-enhanced Raman scattering (SERS) was used as the signal detection method. Sensing mechanisms were based on hybridization of target sequences and DNA probes, resulting in a distance change between SERS reporters and the nanochip's plasmonic-active surface. As the field intensity of the surface plasmon decays exponentially as a function of distance, the distance change in turn affects SERS signal intensity, thus indicating the presence and capture of the target sequences. Our techniques were single-step DNA detection techniques. Target sequences were detected by simple delivery of sample solutions onto DNA probe-functionalized nanochips and measuring the SERS signal after appropriate incubation times. Target sequence labeling or washing to remove unreacted components was not required, making the techniques simple, easy-to-use, and cost-effective. The usefulness of the nanochip platform-based techniques for medical diagnostics was illustrated by the detection of host genetic biomarkers for respiratory viral infection and of the dengue virus gene.

DNA bioassay-on-chip using SERS detection for dengue diagnosis.

A novel DNA bioassay-on-chip using surface-enhanced Raman scattering (SERS) on a bimetallic nanowave chip is presented. In this bioassay, SERS signals were measured after a single reaction on the chip's surface without any washing step, making it simple-to-use and reducing the reagent cost. Using the technique, specific oligonucleotide sequences of the dengue virus 4 were detected.

Multiplex detection of disease biomarkers using SERS molecular sentinel-on-chip.

Developing techniques for multiplex detection of disease biomarkers is important for clinical diagnosis. In this work, we have demonstrated for the first time the feasibility of multiplex detection of genetic disease biomarkers using the surface-enhanced Raman scattering (SERS)-based molecular sentinel-on-chip (MSC) diagnostic technology. The molecular sentinel (MS) sensing mechanism is based upon the decrease of SERS intensity when Raman labels tagged at 3'-ends of MS nanoprobes are physically displaced from the nanowave chip's surface upon DNA hybridization. The use of bimetallic layer (silver and gold) for the nanowave fabrication was investigated. SERS measurements were performed immediately following a single hybridization reaction between the target single-stranded DNA sequences and the complementary MS nanoprobes immobilized on the nanowave chip without requiring target labeling (i.e., label-free), secondary hybridization, or post-hybridization washing, thus shortening the assay time and reducing cost. Two nucleic acid transcripts, interferon alpha-inducible protein 27 and interferon-induced protein 44-like, are used as model systems for the multiplex detection concept demonstration. These two genes are well known for their critical role in host immune response to viral infection and can be used as molecular signature for viral infection diagnosis. The results indicate the potential of the MSC technology for nucleic acid biomarker multiplex detection.

Chronic Effects of e-Cigarette Aerosol Inhalation on Macular Perfusion Assessed Using OCT Angiography.

Purpose: To determine whether there are significant differences in the microvasculature and central retinal thickness (CRT) between e-cigarette users (user group) and age-matched nonusers (control group) using optical coherence tomography angiography (OCTA). Methods: In this prospective cross-sectional observational study, OCTA images were acquired of 52 eyes of 26 users and 25 eyes of 25 age-matched nonusers. Daily e-cigarette users with no ocular history were identified from provider information in the electronic medical record. A custom algorithm was used to calculate the foveal avascular zone (FAZ), vessel area density (VAD), and vessel length density (VLD). OCT software was used to calculate the foveal, superior, inferior, nasal, and temporal CRT. Generalized estimating equations using the Z-statistic were used to determine how the FAZ, VAD, VLD, and CRT parameters varied between groups and to assess the differential contribution of descriptive data in the user group. Results: No statistically significant difference was found between the user group and control group in the FAZ, superficial vascular complex (SVC) VAD, SVC VLD, or deep vascular complex (DVC) VAD. A statistically significant difference was found for DVC VLD (P = .002), with the user group having a slightly higher VLD on average. Superior, temporal, and inferior inner macular thicknesses were significantly thinner in the user group (P = .038, P = .012, and P = .035, respectively). Conclusions: Significant negative differences were found in CRT measures but not in retinal microvasculature parameters between e-cigarette users and nonusers. Decreased inferior, temporal, and superior inner macular thickness in e-cigarette users may show an early chronic structural effect that warrants further assessment of retinal effects as this population ages and continues to use e-cigarettes.

Label-free DNA biosensor based on SERS Molecular Sentinel on Nanowave chip.

Development of a rapid, cost-effective, label-free biosensor for DNA detection is important for many applications in clinical diagnosis, homeland defense, and environment monitoring. A unique label-free DNA biosensor based on Molecular Sentinel (MS) immobilized on a plasmonic 'Nanowave' chip, which is also referred to as a metal film over nanosphere (MFON), is presented. Its sensing mechanism is based upon the decrease of the surface-enhanced Raman scattering (SERS) intensity when Raman label tagged at one end of MS is physically separated from the MFON's surface upon DNA hybridization. This method is label-free as the target does not have to be labeled. The MFON fabrication is relatively simple and low-cost with high reproducibility based on depositing a thin shell of gold over close-packed arrays of nanospheres. The sensing process involves a single hybridization step between the DNA target sequences and the complementary MS probes on the Nanowave chip without requiring secondary hybridization or posthybridization washing, thus resulting in rapid assay time and low reagent usage. The usefulness and potential application of the biosensor for medical diagnostics is demonstrated by detecting the human radical S-adenosyl methionine domain containing 2 (RSAD2) gene, a common inflammation biomarker.

Septins promote caspase activity and coordinate mitochondrial apoptosis.

Apoptosis is a form of regulated cell death essential for tissue homeostasis and embryonic development. Apoptosis also plays a key role during bacterial infection, yet some intracellular bacterial pathogens (such as Shigella flexneri, whose lipopolysaccharide can block apoptosis) can manipulate cell death programs as an important survival strategy. Septins are a component of the cytoskeleton essential for mitochondrial dynamics and host defense, however, the role of septins in regulated cell death is mostly unknown. Here, we discover that septins promote mitochondrial (i.e., intrinsic) apoptosis in response to treatment with staurosporine (a pan-kinase inhibitor) or etoposide (a DNA topoisomerase inhibitor). Consistent with a role for septins in mitochondrial dynamics, septins promote the release of mitochondrial protein cytochrome c in apoptotic cells and are required for the proteolytic activation of caspase-3, caspase-7, and caspase-9 (core components of the apoptotic machinery). Apoptosis of HeLa cells induced in response to infection by S. flexneri ΔgalU (a lipopolysaccharide mutant unable to block apoptosis) is also septin-dependent. In vivo, zebrafish larvae are significantly more susceptible to infection with S. flexneri ΔgalU (as compared to infection with wildtype S. flexneri), yet septin deficient larvae are equally susceptible to infection with S. flexneri ΔgalU and wildtype S. flexneri. These data provide a new molecular framework to understand the complexity of mitochondrial apoptosis and its ability to combat bacterial infection.

Rapid and Portable Quantification of HIV RNA via a Smartphone-enabled Digital CRISPR Device and Deep Learning.

For the 28.2 million people in the world living with HIV/AIDS and receiving antiretroviral therapy, it is crucial to monitor their HIV viral loads with ease. To this end, rapid and portable diagnostic tools that can quantify HIV RNA are critically needed. We report herein a rapid and quantitative digital CRISPR-assisted HIV RNA detection assay that has been implemented within a portable smartphone-based device as a potential solution. Specifically, we first developed a fluorescence-based reverse transcription recombinase polymerase amplification (RT-RPA)-CRISPR assay for isothermally and rapidly detecting HIV RNA at 42 °C in < 30 min. When realized within a commercial stamp-sized digital chip, this assay yields strongly fluorescent digital reaction wells corresponding to HIV RNA. The isothermal reaction condition and the strong fluorescence in the small digital chip unlock compact thermal and optical components in our device, allowing us to engineer a palm-size (70 × 115 × 80 mm) and lightweight (< 0.6 kg) device. Further leveraging the smartphone, we wrote a custom app to control the device, perform the digital assay, and acquire fluorescence images throughout the assay time. We additionally trained and verified a Deep Learning-based algorithm for analyzing fluorescence images and detecting strongly fluorescent digital reaction wells. Using our smartphone-enabled digital CRISPR device, we were able to detect 75 copies of HIV RNA in 15 min and demonstrate the potential of our device toward convenient monitoring of HIV viral loads and combating the HIV/AIDS epidemic.

Ensemble learning using traditional machine learning and deep neural network for diagnosis of Alzheimer's disease.

In recent years, Alzheimer's disease (AD) diagnosis using neuroimaging and deep learning has drawn great research attention. However, due to the scarcity of training neuroimaging data, many deep learning models have suffered from severe overfitting. In this study, we propose an ensemble learning framework that combines deep learning and machine learning. The deep learning model was based on a 3D-ResNet to exploit 3D structural features of neuroimaging data. Meanwhile, Extreme Gradient Boosting (XGBoost) machine learning was applied on a voxel-wise basis to draw the most significant voxel groups out of the image. The 3D-ResNet and XGBoost predictions were combined with patient demographics and cognitive test scores (Mini-Mental State Examination (MMSE) and Clinical Dementia Rating (CDR)) to give a final diagnosis prediction. Our proposed method was trained and validated on brain MRI brain images of the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. During the training phase, multiple data augmentation methods were employed to tackle overfitting. Our test set contained only baseline scans, i.e., the first visit scans since we aimed to investigate the ability of our approach in detecting AD during the first visit of AD patients. Our 5-fold cross-validation implementation achieved an average AUC of 100% during training and 96% during testing. Using the same computer, our method was much faster in scoring a prediction, approximately 10 min, than feature extraction-based machine learning methods, which often take many hours to score a prediction. To make the prediction explainable, we visualized the brain MRI image regions that primarily affected the 3D-ResNet model's prediction via heatmap. Lastly, we observed that proper generation of test sets was critical to avoiding the data leakage issue and ensuring the validity of results.

Filtration-assisted magnetofluidic cartridge platform for HIV RNA detection from blood.

The ability to detect and quantify HIV RNA in blood is essential to sensitive detection of infections and monitoring viremia throughout treatment. Current options for point-of-care HIV diagnosis (i.e. lateral flow rapid tests) lack sensitivity for early detection and are unable to quantify viral load. HIV RNA diagnostics typically require extensive pre-processing of blood to isolate plasma and extract nucleic acids, in addition to expensive equipment for conducting nucleic acid amplification and fluorescence detection. Therefore, molecular HIV diagnostics is still mainly limited to clinical laboratories and there is an unmet need for high sensitivity point-of-care screening and at-home HIV viral load quantification. In this work, we outline a streamlined workflow for extraction of plasma from whole blood coupled with HIV RNA extraction and quantitative polymerase chain reaction (qPCR) in a portable magnetofluidic cartridge platform for use at the point-of-care. Viral particles were isolated from blood using manual filtration through a 3D-printed filter module in seconds followed by automated nucleic acid capture, purification, and transfer to qPCR using magnetic beads. Both nucleic acid extraction and qPCR were integrated within cartridges using compact instrumentation consisting of a motorized magnet arm, miniaturized thermocycler, and image-based fluorescence detection. We demonstrated detection down to 1000 copies of HIV viral particles from whole blood in <30 minutes.

A Portable Droplet Magnetofluidic Device for Point-of-Care Detection of Multidrug-Resistant Candida auris.

Candida auris is an emerging multidrug-resistant fungal pathogen that can cause severe and deadly infections. To date, C. auris has spurred outbreaks in healthcare settings in thirty-three countries across five continents. To control and potentially prevent its spread, there is an urgent need for point-of-care (POC) diagnostics that can rapidly screen patients, close patient contacts, and surveil environmental sources. Droplet magnetofluidics (DM), which leverages nucleic acid-binding magnetic beads for realizing POC-amenable nucleic acid detection platforms, offers a promising solution. Herein, we report the first DM device-coined POC.auris-for POC detection of C. auris. As part of POC.auris, we have incorporated a handheld cell lysis module that lyses C. auris cells with 2 min hands-on time. Subsequently, within the palm-sized and automated DM device, C. auris and control DNA are magnetically extracted and purified by a motorized magnetic arm and finally amplified via a duplex real-time quantitative PCR assay by a miniaturized rapid PCR module and a miniaturized fluorescence detector-all in ≤30 min. For demonstration, we use POC.auris to detect C. auris isolates from 3 major clades, with no cross reactivity against other Candida species and a limit of detection of ∼300 colony forming units per mL. Taken together, POC.auris presents a potentially useful tool for combating C. auris.

Automatic Acne Object Detection and Acne Severity Grading Using Smartphone Images and Artificial Intelligence.

Skin image analysis using artificial intelligence (AI) has recently attracted significant research interest, particularly for analyzing skin images captured by mobile devices. Acne is one of the most common skin conditions with profound effects in severe cases. In this study, we developed an AI system called AcneDet for automatic acne object detection and acne severity grading using facial images captured by smartphones. AcneDet includes two models for two tasks: (1) a Faster R-CNN-based deep learning model for the detection of acne lesion objects of four types, including blackheads/whiteheads, papules/pustules, nodules/cysts, and acne scars; and (2) a LightGBM machine learning model for grading acne severity using the Investigator's Global Assessment (IGA) scale. The output of the Faster R-CNN model, i.e., the counts of each acne type, were used as input for the LightGBM model for acne severity grading. A dataset consisting of 1572 labeled facial images captured by both iOS and Android smartphones was used for training. The results show that the Faster R-CNN model achieves a mAP of 0.54 for acne object detection. The mean accuracy of acne severity grading by the LightGBM model is 0.85. With this study, we hope to contribute to the development of artificial intelligent systems to help acne patients better understand their conditions and support doctors in acne diagnosis.

Plasmonic assay for amplification-free cancer biomarkers detection in clinical tissue samples.

Developing countries have seen a rise in cancer incidence and are projected to harbor three-quarters of all cancer-related mortality by 2030. While disproportionally affected by the burden of cancer, these regions are ill-equipped to handle the diagnostic caseload. The low number of trained pathologists per capita results in delayed diagnosis and treatment, ultimately contributing to increased mortality rates. To address this issue, we developed a point-of-care (POC) plasmonic assay for direct detection of cancer as an alternative to pathological review. Whereas our assay has general applicability in many cancer diagnoses that involve tissue biopsies, we use head and neck cancer (HNC) as a model system because these tumors are increasingly prevalent in lower-income and underserved regions, due to risk factors such as smoking, drinking, and viral infection. Our method uses surface-enhanced Raman scattering (SERS) to detect unique RNA biomarkers from human biopsy samples without the need for complex target amplification machinery (e.g., PCR), making it time and resource-efficient. Unlike previous studies that required target amplification, this work represents a significant advance for HNC diagnosis directly in clinical samples, using only our SERS-based assay for RNA biomarkers. In this study, we tested our assay on 20 clinical samples, demonstrating the accuracy of the method in the diagnosis of head and neck squamous cell carcinoma. We reported sensitivity of 100% and specificity of 97%. Furthermore, we used a handheld Raman device to read the results in order to illustrate the applicability of our method for POC diagnosis of cancer in low-resource settings.