Gain In Sensitivity Research Articles

800-Gb/s/carrier WDM Coherent Transmission Over 2000 km Based on Truncated PS-64QAM Utilizing MIMO Volterra Equalizer

We provided a feasible solution for 800-Gb/s per carrier long-distance wavelength division multiplexing (WDM) coherent optical fibre transmission based on offline DSP. We experimentally demonstrate 5-carrier 125-GHz-spacing WDM coherent optical fibre transmission achieving 2000-km transmission distance and 800-Gb/s net bit rate per carrier. We utilized 100-Gbaud truncated probabilistically shaped (TPS) polarization-multiplexed 64QAM format in this work. In terms of advanced components, the high-speed digital-to-analog converter with 35-GHz 3 dB-bandwidth and Raman-amplified ultra-large-area fibre were utilized to achieve higher baud-rate and longer transmission distance. In terms of modulation format, we changed PS to TPS to obtain a more suitable shaping depth, and make the shaped signal better suited to the DSP process of coherent optical communication, including constant modulus algorithm (CMA) equalizer. Besides, our results show that the TPS-64QAM signal with 5-bit/symbol/polarization entropy outperforms the standard-32QAM with a same bit rate by around 1 dB sensitivity gain. Furthermore, we proposed a multiple-input multiple-output (MIMO) Volterra equalizer (VE), which consists of a 1^st-order 2 × 2 MIMO complex-valued widely linear (WL) equalizer and a 2^nd-order real-valued nonlinear equalizer. Our proposed MIMO VE can effectively resolve the impact of In-phase/quadrature (IQ) imbalance and IQ skew, and compensate linear and nonlinear impairments at the same time. We also utilized partial kernel VE to reduce the complexity by retaining part of the 2^nd-order kernels.

The use of pseudo-MRM for a sensitive and selective detection and quantification of polycyclic aromatic compounds by tandem mass spectrometry.

Multiple Reaction Monitoring (MRM) is a sensitive and selective detection mode for target trace-level analysis. However, it requires the fragmentation of labile bonds which are not present in molecules such as Polycyclic Aromatic Hydrocarbons (PAHs) and their heterocyclic derivatives (PANHs, PASHs). We present the application of an alternative tandem mass spectrometry (MS/MS) mode called "pseudo-MRM" for the GCMS/MS analysis of Polycyclic Aromatic Compounds (PACs). This mode is based on the monitoring of transitions with no mass loss between the precursor and the product ion. Pseudo-MRM peak areas were compared with those of classic MRM on three different mass spectrometers: two triple quadrupoles and an ion trap. For all non-polar PACs studied here (PAHs, PANHs and PASHs), the pseudo-MRM transition was always the most intense. The classic MRM transitions exhibited peak areas 2 to 5 times lower. On the contrary, for the functionalized PACs (oxygenated and nitrated PAHs), classic MRM was favored over pseudo-MRM. These observations were confirmed on two triple quadrupoles (QqQs), and the real-world applicability of pseudo-MRM on QqQs was validated by the successful analysis of Diesel PM. However, a comparison with an ion trap showed that pseudo-MRM was never favored on that instrument, which caused fragmentation of non-polar PACs in MS/MS. The results of this study show an important gain in sensitivity when using pseudo-MRM instead of MRM for non-polar PACs on QqQ instruments. The selectivity of MRM is preserved in pseudo-MRM by applying non-zero collision energies to which only these non-polar PACs are resistant, not the isobaric interferences. No interference issue was observed when analyzing Diesel PM, a complex matrix, with our pseudo-MRM method. Therefore, we advise for a broader use of this MS/MS mode for trace-level determination of non-polar PAHs.

Does social capital alleviate financing constraints? A study of China’s creative economy

ABSTRACT This study investigates the effect of social capital on financial constraints of Chinese cultural and creative enterprises (CCEs). We disaggregate a firm’s social capital into professional, political, and commercial capital. Our empirical results show that professional social capital can significantly reduce CCEs’ investment-cash flow sensitivity and help gain bank loans, thereby alleviating CCEs’ financial constraints. However, political and commercial social capital do not lessen financial constraints. We also find evidence that intellectual property protection acts to amplify the negative relation between professional social capital and financial constraints. These results are robust to alternative measures of key constructs and model specifications.

Development of Prototype Fractional Order Controller for a Grid-Tied Photovoltaic System

This paper presents design of a robust fractional order proportional-integral (FO-PI) prototype controller for a grid connected photovoltaic (PV) system. The robustness in face of parameter variation is handled by designing a suitable controller considering sensitivity function, gain margin and phase margin for both inner and outer control loops. As FO-PI controller is not yet available in physical form and due to complexity involved in grid-tied PV system, we propose an alternative way of hardware implementation of inner loop FO-PI controller by using fractor through RC ladder network. Firstly, an emulator circuit of the inner current loop comprising the FO-PI controller (including fractor and the proportional gain) along with the transfer function of inverter and filter is developed and the response is verified through a real-time simulator (OPAL-RT). Finally, the FO-PI part of the emulator circuit is interfaced with the PC target to form a hardware-in-loop (HIL) by employing OPAL-RT. The performance of the complete system is verified through OPAL-RT simulator. The result shows the improved robustness by the proposed control design method.

Fixed-Point Analysis and FPGA Implementation of Deep Neural Network Based Equalizers for High-Speed PON

Adeep neural network based equalizer is proposed to mitigate the intersymbol interference observed in next generation high speed passive optical network (PON) links. The DNN based equalizer is shown to outperform the best known conventional equalizer, the maximum likelihood sequence estimator (MLSE) both in back-back and through fiber experiments. To reduce the hardware complexity of DNN based equalizer for PON systems, we investigate the use of embedded parallelization within a DNN structure having multiple symbol outputs from one DNN. We further investigate using a classification output stage with cross entropy cost to perform joint decision on multiple symbol outputs and demonstrated that the sensitivity gain of such scheme over regression output. To understand the complexity of hardware implementation, the fixed-point DNN based equalizers are developed and implemented in FPGA. The impact of fixed-point resolution on the receiver sensitivity and hardware resource utilization in FPGA implementation is analyzed and reported in detail. We show that a reduction of over 40% in LUTs (look up table) utilization is possible by reducing the DNN's weight resolution from 8-bit to 4-bit while incurring a small penalty in receiver sensitivity.

Perinatal low-fat dietary intervention affects glucose metabolism in female adult and aging offspring.

Diabetes confers a high risk of developing poor health in later life in women. Based on the Developmental Origins of Health and Disease theory, the present study was undertaken to investigate the efficacy of perinatal fat restriction in maternal high-fat-exposed female offspring to maintain glucose homeostasis in later life between adulthood and aging. Low-fat dietary intervention during either gestation or lactation was performed using a high-fat diet-induced maternal obesity mouse model (HFD mice). Physiological metabolic parameters, including body weight and serum levels of total cholesterol and triglycerides, were monitored. Glucose tolerance test and insulin sensitivity test were performed in 12- and 70-week-old offspring. Insulin-positive islet cells were also observed using immunohistochemical staining. HFD significantly induced abnormal weight gain, hyperlipidemia and impairment of both glucose tolerance and insulin sensitivity in offspring. Standard diet intake after weaning improved weight gain, serum total cholesterol level and glucose tolerance, but not insulin sensitivity, in 70-week-old offspring. Only perinatal fat restriction during both gestation and lactation, followed by standard food intake for the rest of their life, provided adequate efficacy to restore insulin sensitivity in aging female progeny. Perinatal low-fat intervention may prevent deterioration of glucose metabolism. To improve the health status over a female's lifespan, appropriate nutritional intervention during the early developmental stage may reset the disease trajectory and prevent the onset and development of diabetes. Geriatr Gerontol Int 2022; 22: 441-448.

Developing Robust Flood Susceptibility Model with Small Numbers of Parameters in Highly Fertile Regions of Northwest Bangladesh for Sustainable Flood and Agriculture Management

The present study intends to improve the robustness of a flood susceptibility (FS) model with a small number of parameters in data-scarce areas, such as northwest Bangladesh, by employing machine learning-based sensitivity analysis and an analytical hierarchy process (AHP). In this study, the nine most relevant flood elements (such as distance from the river, rainfall, and drainage density) were chosen as flood conditioning variables for modeling. The FS model was produced using AHP technique. We used an empirical and binormal receiver operating characteristic (ROC) curves for validating the models. We performed Sensitivity analyses using a random forest (RF)-based mean Gini decline (MGD), mean decrease accuracy (MDA), and information gain ratio to find out the sensitive flood conditioning variables. After performing sensitivity analysis, the least sensitivity variables were eliminated. We re-ran the model with the rest of the parameters to enhance the model’s performance. Based on previous studies and the AHP weighting approach, the general soil type, rainfall, distance from river/canal (Dr), and land use/land cover (LULC) had higher factor weights of 0.22, 0.21, 0.19, and 0.15, respectively. The FS model without sensitivity and with sensitivity performed well in the present study. According to the RF-based sensitivity and information gain ratio, the most sensitive factors were rainfall, soil type, slope, and elevation, while curvature and drainage density were less sensitive parameters, which were excluded in re-running the FS model with just vital parameters. Using empirical and binormal ROC curves, the new FS model yields higher AUCs of 0.835 and 0.822, respectively. It is discovered that the predicted model’s robustness may be maintained or increased by removing less relevant factors. This study will aid decision-makers in developing flood management plans for the examined region.

Combined Diagnostic Accuracy of Diffusion and Perfusion MR Imaging to Differentiate Radiation-Induced Necrosis from Recurrence in Glioblastoma.

We aimed to use quantitative values derived from perfusion and diffusion-weighted MR imaging (PWI and DWI) to differentiate radiation-induced necrosis (RIN) from tumor recurrence in Glioblastoma (GBM) and investigate the best parameters for improved diagnostic accuracy and clinical decision-making. Methods: A retrospective analysis of follow-up MRI with new enhancing observations was performed in histopathologically confirmed subjects of post-treated GBM, who underwent re-surgical exploration. Quantitative estimation of rCBV (relative cerebral blood volume) from PWI and three methods of apparent diffusion coefficient (ADC) estimation were performed, namely ADC R1 (whole cross-sectional area of tumor), ADC R2 (only solid enhancing lesion), and ADC R3 (central necrosis). ROC curve and logistic regression analysis was completed. A confusion matrix table created using Excel provided the best combination parameters to ameliorate false-positive and false-negative results. Results: Forty-four subjects with a mean age of 46 years (range, 19–70 years) underwent re-surgical exploration with RIN in 28 (67%) and recurrent tumor in 16 (33%) on histopathology. rCBV threshold of >3.4 had the best diagnostic accuracy (AUC = 0.93, 81% sensitivity and 89% specificity). A multiple logistic regression model showed significant contributions from rCBV (p < 0.001) and ADC R3 (p = 0.001). After analysis of confusion matrix ADC R3 > 2032 × 10−6 mm2 achieved 100% specificity with gain in sensitivity (94% vs. 56%). Conclusions: A combination of parameters had better diagnostic performance, and a stepwise combination of rCBV and ADC R3 obviated unnecessary biopsies in 10% (3/28), leading to improved clinical decision-making.

Lower beta-1,3-D-glucan testing cut-offs increase sensitivity for non-albicans Candida species bloodstream infections.

Fungal biomarkers support early diagnosis of invasive fungal infections. In this study, we evaluated the impact of a recent update to the manufacturer-recommended cut-off for beta-1,3-D-glucan (BDG) testing (Fujifilm Wako BDG assay) on sensitivity and specificity for the detection of candidemia. Additionally, we compared the performance with tests for Candida antigen (Ag by Serion ELISA antigen Candida, Virion\Serion) and anti-mannan antibodies (Ab by Hemkit Candida IHA, Ravo Diagnostika). Sera of 82 patients with candidemia, which were sampled with a maximum distance of ±14days from the date of sampling of the corresponding positive blood cultures, were retrospectively analysed for BDG, Ag and Ab. Results of BDG testing were compared with results from sera of 129 patients with candidemia from a different hospital. Sensitivity of BDG testing (47%) was higher than for Ag (17%) or Ab (20%). By combining Ag and Ab testing, sensitivity was raised to 32%. Lowering the cut-off of BDG from 11pg/ml to the newly recommended cut-off of 7pg/ml resulted in a significant increase in sensitivity (47% vs 58%, p=.01 and 63% vs 71% p<.01). At both centres, the increase was significant in NAC but not in C.albicans candidemia. No significant effects on specificity were observed. BDG testing outperformed Ag and Ab testing and its combination. Lowering the BDG cut-off had no significant impact on specificity. The increase in sensitivity can be mainly attributed to a gain in sensitivity for non-albicans Candida species bloodstream infections.

Dynamic Nuclear Polarization Solid-State NMR Spectroscopy for Materials Research

Solid-state nuclear magnetic resonance (NMR) spectroscopy has increasingly been used for materials characterization as it enables selective detection of elements of interest, as well as their local structure and dynamic properties. Nevertheless, utilization of NMR is limited by its inherent low sensitivity. The development of dynamic nuclear polarization (DNP) approaches, which provide enormous sensitivity gain in NMR through the transfer of polarization from electron spins, has transformed the application of solid-state NMR in materials science. In this review, we outline the opportunities for materials characterization that DNP has opened up. We describe the main DNP mechanisms available, their implementation, and the kinds of insight they can provide across different materials classes, from surfaces and interfaces to defects in the bulk of solids. Finally, we discuss the current limitations of the approach and provide an outlook on future developments for DNP-enhanced NMR spectroscopy in materials science.

Quasi-2D Perovskite Crystalline Layers for Printable Direct Conversion X-Ray Imaging.

Polycrystalline perovskite film-based X-ray detector is an appealing technology for assembling large scale imager by printing methods. However, thick crystalline layer without trap and solvent residual is challenging to fabricate. Here, the authors report a solution method to produce high quality quasi-2D perovskite crystalline layers and detectors that are suitable for X-ray imaging. By introducing n-butylamine iodide into methylammonium lead iodide precursor and coating at elevated temperatures, compact and crystalline layers with exceptional uniformity are obtained on both rigid and flexible substrates. Photodiodes built with the quasi-2D layers exhibit a low dark current and stable operation under constant electrical field over 96 h in dark, and over 15 h under X-ray irradiation. The detector responds sensitively under X-ray, delivering a high sensitivity of 1214µCGyair -1 cm-2 and a sensitivity gain is observed when operated under higher fields. Finally, high resolution images are demonstrated using a single pixel device that can resolve 80-200µm features. This work paves the path for printable direct conversion X-ray imager development.

The Wavelength-Shifting Optical Module.

The Wavelength-shifting Optical Module (WOM) is a novel photosensor concept for the instrumentation of large detector volumes with single-photon sensitivity. The key objective is to improve the signal-to-noise ratio, which is achieved by decoupling the photosensitive area of a sensor from the cathode area of its photomultiplier tube (PMT). The WOM consists of a transparent tube with two PMTs attached to its ends. The tube is coated with wavelength-shifting paint that absorbs ultraviolet photons with nearly 100% efficiency. Depending on the environment, e.g., air (ice), up to 73% (41%) of the subsequently emitted optical photons can be captured by total internal reflection and propagate towards the PMTs, where they are recorded. The optical properties of the paint, the geometry of the tube, and the coupling of the tube to the PMTs have been optimized for maximal sensitivity based on theoretical derivations and experimental evaluations. Prototypes were built to demonstrate the technique and to develop a reproducible construction process. Important measurable characteristics of the WOM are the wavelength-dependent effective area, the transit time spread of detected photons, and the signal-to-noise ratio. The WOM outperforms bare PMTs, especially with respect to the low signal-to-noise ratio with an increase of a factor up to 8.9 in air (5.2 in ice). Since the gain in sensitivity is mostly in the UV regime, the WOM is an ideal sensor for Cherenkov and scintillation detectors.

Perspectives d’utilisation clinique de la tomographie par émission de positons grand-champ (TEP-GC)

In the last decade, PET scanners have benefited from new technologies such as TOF support, PSF correction and SiPM detectors. More recently, the increase of detector number allowed an axial FOV as large as a human body length in total-body PET (TB-PET) innovative systems. Two different scanners called Quadra© and uEXPLORER© have been respectively proposed by Siemens Healthineers and United Imaging Healthcare. These TB-PET have two major advantages over actual systems on the market: a gain in detection sensitivity and a possibility of whole-body dynamic multiple organs analysis [1,2]. More than reducing drastically the acquisition duration and/or injected activities, these systems open up very interesting clinical perspectives to address specific issues encountered in routine clinical practice and thus improve patient care. We aimed in this brief communication to highlight some situations for which a TB-PET might help to settle current clinical issues.

Tillyer Award Lecture: Visual search in noise and natural backgrounds

I will describe evidence for a theory of covert visual search developed within the framework of natural scene statistics and Bayesian statistical decision theory. The theory is unique in several ways: (1) it directly takes into account the statistical properties of natural images, (2) it takes into account the variation in neural processing with retinal location, as well as other known properties of the visual system, and hence contains almost no free parameters, and (3) it includes a principled attentional mechanism that efficiently allocates sensitivity gain across the visual field. This latter mechanism was discovered in experiments measuring covert search in white-noise backgrounds, where the target could appear anywhere within a large search area. In a separate experiment, target detectability (d') was measured across the visual field when the target location was cued/known. The shape of this "d' map" was consistent with the theory. The overall performance in the covert search task was also predicted quite well from this d' map, with no free parameters, assuming parallel unlimited-capacity processing. However, paradoxically, detection accuracy was low in the foveal region, even though it was predicted to be very high. We show that this "foveal neglect" is the expected consequence of efficiently allocating a fixed total attentional sensitivity gain across neurons in visual cortex, rather than across locations in visual space (the traditional assumption). Furthermore, the theory predicts the detailed pattern of covert search performance in the white-noise backgrounds. Finally, I will describe predictions of the theory for search in natural images.

Highly Secure and Reliable 7-Core Fiber Optical OFDM Access System Based on Chaos Encryption Inside Polar Code

A highly secure and reliable optical orthogonal frequency division multiplexing passive optical network (OFDM-PON) based on the chaos encryption inside polar code scheme is proposed. The 3-dimensional Chua's circuit model is adopted to realize chaotic encryption during polar encoding with low complexity. The three chaotic sequences generated by Chua's circuit model are used to encrypt information index bit matrix, frozen bits (during polar encoding) and subcarrier sequence, respectively. The BER and security performance of the optical transmission system are enhanced by our method. A 70 Gb/s polar encoded encrypted 16 quadrature amplitude modulation OFDM signal transmission over a 2 km seven-core fiber is further experimentally demonstrated. The experimental results show that the polar encoded encrypted 16 QAM-OFDM obtains a 2.1 dB gain in receiver sensitivity at a BER of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−3</sup> under the same bit rate compared to the normal 16 QAM-OFDM without FEC. Moreover, the key space of the proposed scheme is 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">84</sup> , which is large enough to guard against any malicious attacks from illegal ONUs effectively. The combined superiority of BER and security performance enable such OFDM-PON based on chaos encryption inside polar code, to have a high prospect of application in low-cost and reliable optical access system.

Efficient Pulsed Dynamic Nuclear Polarization with the X-Inverse-X Sequence.

Pulsed dynamic nuclear polarization (DNP) is a promising new approach to enhancing the sensitivity of high-resolution magic-angle spinning (MAS) NMR. In pulsed DNP, the transfer of polarization from unpaired electrons to nuclei (usually 1H) is induced by a sequence of microwave pulses. Enhancement factors of the thermal 1H polarization are expected to be independent of the magnetic field, and sample heating by absorption of microwave irradiation will be strongly reduced. The development of DNP pulse sequences is still in its infancy. Of the two basic sequences in existence, NOVEL and TOP DNP, the former is, due to an extremely high power requirement, incompatible with high-resolution MAS NMR, while the latter displays a relatively slow transfer of polarization from electrons to 1H. We introduce here a new pulse sequence for DNP of solids, termed X-inverse-X (XiX) DNP. In experiments at 1.2 T, XiX DNP produces, compared to TOP DNP, a 2-fold higher gain in sensitivity. Our data suggest that a faster transfer of polarization from electrons to 1H is behind the superior performance of XiX DNP. Numerical simulations and experiments indicate that microwave pulse lengths can be chosen across a broad range, without loss of efficiency. These findings are a substantial step toward the implementation of pulsed DNP at high magnetic fields.

Flexible sensitive hydrogel sensor with self-powered capability

Flexible sensitive hydrogels gain comprehensive research interests in fabricating human-machine interfacial devices. However, they face a great challenge of incompatibility among high strength, ultra-toughness and good conductivity, and count on an external power supply. In the work, 4-(methacryloyloxy) ethyl-1- phenylene acid (MBA) and 1,4-(5-hexenyloxy) benzene (RM26) are synthesized as supramolecular crosslinker and inserted into polyvinyl alcohol (PVA) hydrogel. A series of conductive hydrogels (PVA-PMR-NaCls) are produced by salt immersion method. The optimal PVA-PMR-NaCl exhibits the toughness of 15.92MJ·m−3, the elongation at break of 600%, the tensile strength of 5.61 MPa and the Young's modulus of 2.16 MPa. It withstands the side effect of saline ions because the defined molecules with flexible chain segment and rigid group intervene. The conductivity for PVA-PMR-NaCl is 3.0 S·m−1 prior to original PVA-NaCl owing to the carboxyl in network. It shows high gauge factor (GF) toward tensile strain, which achieves 6.24 above 200% strain. The PVA-PMR-NaCl sensor presents high sensitivity, stability and quick responsiveness during detecting human movement. Furthermore, the self-powered PVA-PMR-NaCl sensor by assembling with zinc foil and copper foil achieves the open circuit voltage of 0.809 V, and can likewise trace the stretching, bending and compressing movement according to current change. So, it will be a promising candidate as a self-powered multi-functional sensor.

77Se-Enriched Selenoglycoside Enables Significant Enhancement in NMR Spectroscopic Monitoring of Glycan-Protein Interactions.

Detailed investigation of ligand–protein interactions is essential for better understanding of biological processes at the molecular level. Among these binding interactions, the recognition of glycans by lectins is of particular importance in several diseases, such as cancer; therefore, inhibition of glycan-lectin/galectin interactions represents a promising perspective towards developing therapeutics controlling cancer development. The recent introduction of 77Se NMR spectroscopy for monitoring the binding of a selenoglycoside to galectins prompted interest to optimize the sensitivity by increasing the 77Se content from the natural 7.63% abundance to 99%. Here, we report a convenient synthesis of 77Se-enriched selenodigalactoside (SeDG), which is a potent ligand of the medically relevant human galectin-3 protein, and proof of the expected sensitivity gain in 2D 1H, 77Se correlation NMR experiments. Our work opens perspectives for adding isotopically enriched selenoglycans for rapid monitoring of lectin-binding of selenated as well as non-selenated ligands and for ligand screening in competition experiments.

Three-dimensional constellation diagram with a hierarchical level design for multi-core transmission.

In this paper, a novel three-dimensional constellation diagram is proposed for optimizing the spatial coordinates of constellation points using geometric shaping (GS), which improves the constellation figure of merit (CFM) and, as a result, reduces the bit error rate (BER). A three-dimensional carrier-free amplitude phase (CAP) modulated signal transmission at 25.45Gb/s is successfully implemented on a seven-core fiber communication system to verify the performance of the constellation diagram. Compared to the traditional scheme, this method has an average receiver sensitivity gain of 1.4dB at BER ∼1 × 10-3, further, the receiver sensitivity gain difference of 1.3dB between different cores. The experimental results show that this scheme effectively reduces BER without additional communication components, which can be used in short-distance access networks in the near future.

Impact of Generative Modeling for Fundus Image Augmentation With Improved and Degraded Quality in the Classification of Glaucoma

Glaucoma is a heterogeneous group of diseases characterised by cupping of the optic nerve head and visual field damage, starting with a progressive loss of vision that leads to permanent blindness. When diagnosed in time, Glaucoma can be delayed by adequate treatment. More efficient processes for diagnosis are being proposed, and the role of artificial intelligence in the field is growing. This work presents a pipeline to evaluate the impact of generative modelling in Computer-Aided Diagnosis (CADx) of Glaucoma based on Deep Learning, particularly focused on the optic disc region. The methodology relies on transforming retinal fundus images to improve and degrade their quality to augment the training data and assess the diagnostic performance. The objective evaluation of the proposed model based on Generative Adversarial Networks revealed quantitative and qualitative improvements in image quality. To support this, we propose a new model to evaluate the quality of fundus images, which can also be used within the pipeline to reject samples with lower image quality for diagnosis. Its performance surpassed related work, achieving a balanced accuracy of 0.929. Concerning Glaucoma CADx, the results obtained in public datasets point to a considerable gain in Sensitivity, Specificity, and Accuracy, achieving scores of 0.883 (+0.054), 0.957 (+0.019), and 0.931 (+0.031), respectively, after image data augmentation when compared with previous work targeted at offline inference in mobile devices. Considering the restriction of choosing simpler backbone networks that can run on edge devices, our findings support the importance of image quality diversity and realistic augmentation.