Subjective Visual Inspection Research Articles

Quantitative Testing of a SOLO-Based Automated Quality Control Algorithm for Airborne Tail Doppler Radar Data

An automated quality control pre-processing algorithm for removing non-weather radar echoes from airborne Doppler radar data has been developed. The proposed algorithm can significantly reduce the time and experience level required for interactive radar data editing prior to dual-Doppler wind synthesis or data assimilation. As important as reducing the time required and skill level necessary to process an airborne Doppler dataset can be, the quality of the automated analysis is paramount. Retrieved wind data, recovered perturbation pressure data (with associated momentum check values) and correlation coefficients were computed. To quantitatively test the quality of the automated quality control algorithm, spatial Pearson correlation coefficients and momentum check values were computed. Four different (published) Electra Doppler Radar (ELDORA) datasets of convective echoes were used to stress the algorithm. Four distinct threshold levels for data removal in the automated quality control algorithm were applied to each of four ELDORA datasets. The algorithm threshold levels were labeled as follows: extremely low, low, medium, and high. Extremely low algorithm cases were deemed necessary during the data analyses and were added to the low, medium and high cases. A description of each case and the differences in the perturbation pressure momentum check values and correlation coefficients between the interactively edited fields were computed. These comparisons along with a subjective visual inspection show that the automated quality control algorithm can produce an analysis comparable—and in some cases superior—to an interactive analysis when used properly. A key benefit of this algorithm is that the skill level of a relatively inexperienced airborne radar meteorologist may be effectively increased by using the SOLO QC algorithm.

Machine Learning-Based Tomato Fruit Shape Classification System.

Fruit shape significantly impacts the quality and commercial value of tomatoes (Solanum lycopersicum L.). Precise grading is essential to elucidate the genetic basis of fruit shape in breeding programs, cultivar descriptions, and variety registration. Despite this, fruit shape classification is still primarily based on subjective visual inspection, leading to time-consuming and labor-intensive processes prone to human error. This study presents a novel approach incorporating machine learning techniques to establish a robust fruit shape classification system. We trained and evaluated seven supervised machine learning algorithms by leveraging a public dataset derived from the Tomato Analyzer tool and considering the current four classification systems as label variables. Subsequently, based on class-specific metrics, we derived a novel classification framework comprising seven discernible shape classes. The results demonstrate the superiority of the Support Vector Machine model in terms of its accuracy, surpassing human classifiers across all classification systems. The new classification system achieved the highest accuracy, averaging 88%, and maintained a similar performance when validated with an independent dataset. Positioned as a common standard, this system contributes to standardizing tomato fruit shape classification, enhancing accuracy, and promoting consensus among researchers. Its implementation will serve as a valuable tool for overcoming bias in visual classification, thereby fostering a deeper understanding of consumer preferences and facilitating genetic studies on fruit shape morphometry.

Detection of Subtle ECG Changes Despite Superimposed Artifacts by Different Machine Learning Algorithms

Analyzing electrocardiographic (ECG) signals is crucial for evaluating heart function and diagnosing cardiac pathology. Traditional methods for detecting ECG changes often rely on offline analysis or subjective visual inspection, which may overlook subtle variations, particularly in the case of artifacts. In this theoretical, proof-of-concept study, we investigated the potential of five different machine learning algorithms [random forests (RFs), gradient boosting methods (GBMs), deep neural networks (DNNs), an ensemble learning technique, as well as logistic regression] to detect subtle changes in the morphology of synthetically generated ECG beats despite artifacts. Following the generation of a synthetic ECG beat using the standardized McSharry algorithm, the baseline ECG signal was modified by changing the amplitude of different ECG components by 0.01–0.06 mV. In addition, a Gaussian jitter of 0.1–0.3 mV was overlaid to simulate artifacts. Five different machine learning algorithms were then applied to detect differences between the modified ECG beats. The highest discriminatory potency, as assessed by the discriminatory accuracy, was achieved by RFs and GBMs (accuracy of up to 1.0), whereas the least accurate results were obtained by logistic regression (accuracy approximately 10% less). In a second step, a feature importance algorithm (Boruta) was used to discriminate which signal parts were responsible for difference detection. For all comparisons, only signal components that had been modified in advance were used for discretion, demonstrating that the RF model focused on the appropriate signal elements. Our findings highlight the potential of RFs and GBMs as valuable tools for detecting subtle ECG changes despite artifacts, with implications for enhancing clinical diagnosis and monitoring. Further studies are needed to validate our findings with clinical data.

Towards automated field ballast condition evaluation: Field validation of the ballast scanning vehicle capabilities

Ballast degradation can lead to adverse effects such as inadequate drainage, track settlement and reduced lateral stability, which could compromise track safety, daily functionality, and long-term maintenance. Field inspection of ballast for monitoring degradation and functional performance is a challenging task. Current state-of-the-practice methods for evaluating ballast primarily depend on subjective visual inspection, labor-intensive sampling, laboratory sieve analyses or Ground Penetrating Radar (GPR) technology. These methods fall short in providing an in-depth assessment of ballast, specifically in determining the degradation level and aggregate size and shape characteristics at various depths. In this regard, this research developed an innovative ballast investigation platform, the Ballast Scanning Vehicle (BSV), to automate the processes of acquiring detailed ballast inspection data. The BSV utilizes a deep learning-based pipeline for image segmentation to evaluate task-specific metrics such as coarse aggregate gradation, Fouling Index (FI), and continuous track FI depth profiles. This paper provides a detailed overview of the BSV’s functions as well as the different modules of the deep learning-based pipeline. Validation of the BSV’s capabilities was conducted at the Transportation Technology Center (TTC) and is discussed in detail. Based on the field results, the BSV is capable of providing accurate and near real-time evaluation of in-service ballast conditions, serves as a robust means for inspecting long sections of track, and can be used to investigate persistent trouble-spots related to track performance.

A spectral assessment for instant preliminary evaluation of structures after seismic events

This article introduces a simple, efficient spectral-based approach for rapid preliminary seismic assessment of earthquake-affected structures. Performance is validated using data from three highly earthquake-affected structures in New Zealand, where visual inspection with subjective outcomes missed damage, resulting in inaccurate and delayed decisions with significant social and economic losses. The three structures considered include (1) the instrumented Bank of New Zealand (BNZ) building; (2) the un-instrumented Queensgate Mall (QM) Complex; and (3) the un-instrumented Canterbury Television (CTV) building. This study uses these cases to highlight the importance of structural health monitoring (SHM) instrumentation and reliable quantified post-earthquake assessment methods in earthquake-prone areas, where each damaging earthquake and subsequent further-damaging aftershocks demand continuous monitoring to continuously assess damage and life safety risk. The simple, low-cost spectral analyses in this study clearly show the existence of damage and deterioration not fully discovered with standard visual inspection methods. This outcome highlights the importance of sensor networks and SHM instrumentation so quantitative, post-event analysis can rapidly augment and target further, more subjective visual inspection results.

A taxonomic revision of the Palaearctic species of the ant genus Tapinoma Mayr 1861 (Hymenoptera: Formicidae)

A taxonomic revision of the Palaearctic species of the ant genus Tapinoma Emery 1861 is provided. Due to the extreme rarity of discernable characters, the female castes of 64% of the species are not separable by primary visual perception of an expert and are thus truly cryptic. The main working rationale of the revision is Numeric Morphology-Based Alpha-Taxonomy (NUMOBAT) with formation of species hypotheses largely based on exploratory data analyses and checking these hypotheses by discriminant analysis. NUMOBAT data of all considered species comprise 648 worker samples with 2014 individuals and 30,000 primary data. Species are depicted by z-stack imaging in three standard visual positions. Numeric data on 15 phenotypical characters are presented in comparative tables and supplementary verbal descriptions are given. Analysis of nuclear DNA was based on evaluation of 15 microsatellite loci in 274 nest samples. The NUMOBAT-based classification of eight species of the Tapinoma nigerrimum and T. simrothi group which are inseparable by subjective visual inspection was confirmed by nuclear DNA data in 98.3% of 169 samples where both morphological and nuDNA data were available. The three cases of disagreement between NUMOBAT and nuDNA classification were discussed. Excluding Nomina Nuda and unavailable names, the revision recognized 34 available names which divide into 21 valid species, 11 junior synonyms and two names placed under Incertae Sedis. Four species are described as new: Tapinoma hispanicum n. sp., T. insularis n. sp., T. dabashanica n. sp. and T. sichuense n. sp. The following synonymies were stated: T. collina Foerster 1850, T. erraticum platyops Emery 1925 and T. erraticum bononiensis Emery 1925 as synonyms of T. erraticum (Latreille 1798); T. tauridis Emery 1925 and T. breve Emery 1925 as synonyms of T. glabrella (Nylander 1849) that is erected to species level; T. minor Bernard 1945 as synonym of T. simrothi Krausse-Heldrungen 1911; T. simrothi azerbaidzhanica Karavajev 1932 as synonym of Tapinoma karavaievi Emery 1925 that is erected here to species level; T. erraticum ambiguum Emery 1925 as synonym of T. madeirense Forel 1895 and T. emeryanum Kuznetzov-Ugamsky 1927 as synonym of T. sinense Emery 1925. Despite an insufficient separation of the female castes of Tapinoma madeirense and T. subboreale Seifert 2011 by NUMOBAT characters and detection of a hybrid zone in southern France, the heterospecific status of the two taxa was maintained based on sufficient differences in male genitalia and of nuclear DNA over a larger geographic scale. Formica caerulescens Losana 1834 is removed from genus Tapinoma and recognized as member of the subfamily Formicinae.

Experimental Study on Dynamic Characteristics of Damaged Post-Tensioning Concrete Sleepers Using Impact Hammer.

Concrete sleepers in operation are commonly damaged by various internal and external factors, such as poor materials, manufacturing defects, poor construction, environmental factors, and repeated loads and driving characteristics of trains; these factors affect the vibration response, mode shape, and natural frequency of damaged concrete sleepers. However, current standards in South Korea require only a subjective visual inspection of concrete sleepers to determine the damage degree and necessity of repair or replacement. In this study, an impact hammer test was performed on concrete sleepers installed on the operating lines of urban railroads to assess the field applicability of the modal test method, with the results indicating that the natural frequency due to concrete sleeper damage was lower than that of the undamaged state. Furthermore, the discrepancy between the simulated and measured natural frequencies of the undamaged concrete sleeper was approximately 1.87%, validating the numerical analysis result. The natural frequency of the damaged concrete sleepers was lower than that of the undamaged concrete sleeper, and cracks in both the concrete sleeper core and the rail seat had the lowest natural frequency among all the damage categories. Therefore, the damage degrees of concrete sleepers can be quantitatively estimated using measured natural-frequency values.

Optimized deep learning for steel bridge bolt corrosion detection and classification

Corrosion of steel bridge bolts poses a major threat to structural integrity. As a key component, bridge bolts are susceptible to corrosion and detachment, necessitating close monitoring during maintenance. However, the vast number of corroded bolts presents challenges. Conventional detection relies on subjective visual inspection without unified assessment criteria. Fast and accurate localization and classification of bolt corrosion levels thus became a research priority for bridge maintenance. This study puts forward a new two-stage corrosion detection method, integrating Deep Convolutional Generative Adversarial Network (DCGAN) and improved You Only Look Once version 5 (YOLOv5) networks, which enables rapid categorization of bridge bolts by different corrosion extents, providing an intelligent solution for bridge maintenance. First, in the image preprocessing stage, a DCGAN with least squares loss generates simulated corroded bolt images through adversarial learning to expand the dataset. Second, three corrosion levels are defined based on visual features and percentage of surface rust, with this classification scheme applied in dataset annotation. Finally, the annotated images are input into an enhanced YOLOv5 model named YOLOv5-SE for corrosion identification. To improve accuracy, Squeeze-and-Excitation blocks are added after each residual block in the YOLOv5 backbone. Comparative experiments prove YOLOv5-SE's superior performance over YOLOv5 and other models like Faster R-CNN and VGG16. A pytorch system enables model deployment for corrosion detection. The integration of DCGAN simulation and attention-optimized YOLOv5-SE enables precise classification and localization of corroded bolts. This study represented a significant step towards intelligent upgrades in steel bridge maintenance.

Chemotaxonomic classification of closed-related sweet oranges (Citrus sinensis L.) using principal component and cluster analysis of color and bi-oactive compounds in the orange peels

Chemotaxonomy is a systematic approach to plant classification based on discernible variations and similarities in their biochemical compositions, with a particular emphasis on secondary metabolites that often exhibit specificity within taxonomically related plant species. In the context of our investigation, we conducted a comprehensive chemotaxonomic analysis of sweet oranges utilized in concentrated juice production. This involved a thorough examination of antioxidant activities, characterization of flavonoid and phenolic profiles, UV-visible spectra scanning, and the implementation of color analysis. The results from cluster analysis demonstrated the superior efficacy of color analysis in discerning subtle distinctions among orange varieties. Further refinement in the analytical process requires the application of supplementary techniques after an exhaustive and precise extraction facilitated by organic solvents. Notably, all employed methodologies successfully discriminated between distinct orange types. While the compilation of extensive datasets presents inherent challenges, the incorporation of Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) enhances the interpretability of the obtained results. Considering that existing methodologies for assessing fresh fruit quality rely on subjective human visual inspection or orange juice analysis by ratio, which may lack precision in quantifying chemical constituents, there exists a growing demand for expedited techniques that can reliably ensure the quality of orange juice within the industry.

DEEP LEARNING-BASED CLASSIFICATION FOR MELANOMA DETECTION USING XCEPTIONNET AND DENSENET121

Melanoma, the deadliest form of skin cancer, poses a significant public health challenge worldwide. Early detection is crucial for improving patient outcomes, yet traditional methods often rely on subjective visual inspection by dermatologists, leading to variability and delay in diagnosis. This study presents a deep learningbased classification approach for melanoma detection, utilizing the XceptionNet and DenseNet121 convolutional neural network (CNN) architectures. Leveraging a large dataset of dermoscopic images, the proposed model is trained to automatically identify malignant melanomas from benign lesions with high accuracy. XceptionNet and DenseNet121 are employed as feature extractors, capturing intricate patterns and features from skin lesion images. Transfer learning techniques are utilized to fine-tune the pretrained models on the melanoma dataset, enhancing classification performance. Extensive experimentation and evaluation on benchmark datasets demonstrate the superior performance of the proposed approach compared to traditional methods and standalone CNN architectures. The deep learning-based classification model holds promise for aiding dermatologists in early melanoma detection, potentially reducing diagnostic variability and improving patient outcomes.

Combining unsupervised deep learning and Monte Carlo dropout for seismic data reconstruction and its uncertainty quantification

Many methods, such as multichannel singular spectrum analysis (MSSA) and deep seismic prior (DSP), have been developed for seismic data reconstruction, but they do not quantify the uncertainty of reconstructed traces, relying on the subjective visual inspection of results. Our goal is to quantify the reconstructed uncertainty while recovering missing traces. We develop a framework including an unsupervised deep-learning-based seismic data reconstruction method and the existing Monte Carlo dropout method to achieve this goal. The only information required by our framework is the original incomplete data. A convolutional neural network trained on the original nonmissing traces can simultaneously denoise and reconstruct seismic data. For uncertainty quantification, the Monte Carlo dropout method treats the well-known dropout technique as Bayesian variational inference. This refers to the fact that the dropout technique can be regarded as an approximation to the probabilistic Gaussian process and thus can be used to obtain an approximate distribution (Bernoulli variational distribution) of the posterior distribution. The reconstructed result and uncertainty of the trained model are yielded through multiple Monte Carlo dropout simulations. The analysis of the reconstructed uncertainty quantifies the confidence to use reconstructed traces. Tests on synthetic and field data illustrate that our framework outperforms the MSSA and DSP methods on reconstructed accuracy and quantifies the reconstructed uncertainty as an objective benchmark to guide decision making.

A new local structural similarity fusion-based thresholding method for homomorphic ultrasound image despeckling in NSCT domain

Ultrasound is a diagnostic imaging technique to detect various illnesses related to the body's internal organs. It is a non-invasive, safe, cost-effective, and simple imaging modality that uses high-frequency sound waves and their reflection from internal organs. However, a key drawback of using ultrasound imaging tools is that they are often contaminated by multiplicative speckle noise, which can reduce the image resolution and contrast and complicates the diagnosis of the tiny structure of lesions. Extensive studies have shown that eliminating speckle noise is a necessary pre-processing task for computer-assisted ultrasound image systems. This paper proposes a novel method for denoising, weighted thresholding, and fusion of ultrasound images in the Non-Subsampled Contourlet Transformation (NSCT) domain. The method utilizes (a) The Bayesian and Bivariate shrinkage rule in the NSCT domain, (b) A new local structural similarity-fusion-based homomorphic ultrasound image despeckling method in the NSCT domain, and (c) A weights estimation approach based on a bilateral method that helps improve noise suppression and edge preservation results. The performance of the proposed algorithm is evaluated in terms of mean squared error (MSE), peak signal-to-noise ratio, structural similarity index (SSIM), speckle suppression index, signal-to-MSE ratio, and speckle signal-to-noise ratio. Extensive computer simulation results show that the presented method performs better in subjective visual inspection than existing state-of-the-art despeckling techniques.

Detection and differentiation of herbicide stresses in roses by Raman spectroscopy.

Herbicide application is a critical component of modern horticulture. Misuse of herbicides can result in damage to economically important plants. Currently, such damage can be detected only at symptomatic stages by subjective visual inspection of plants, which requires substantial biological expertise. In this study, we investigated the potential of Raman spectroscopy (RS), a modern analytical technique that allows sensing of plant health, for pre-symptomatic diagnostics of herbicide stresses. Using roses as a model plant system, we investigated the extent to which stresses caused by Roundup (Glyphosate) and Weed-B-Gon (2, 4-D, Dicamba and Mecoprop-p (WBG), two of the most commonly used herbicides world-wide, can be diagnosed at pre- and symptomatic stages. We found that spectroscopic analysis of rose leaves enables ~90% accurate detection of Roundup- and WBG-induced stresses one day after application of these herbicides on plants. Our results also show that the accuracy of diagnostics of both herbicides at seven days reaches 100%. Furthermore, we show that RS enables highly accurate differentiation between the stresses induced by Roundup- and WBG. We infer that this sensitivity and specificity arises from the differences in biochemical changes in plants that are induced by both herbicides. These findings suggest that RS can be used for a non-destructive surveillance of plant health to detect and identify herbicide-induced stresses in plants.

Molecular classification of cancer site of origin and lineage.

3116 Background: Histopathologic assessment has been the primary modality for the diagnosis of human cancers since the 19th century, and to this day remains the mainstay of diagnosis, risk stratification and staging. While the field has made significant advances, the “art of pathology” relies heavily on subjective visual inspection, with significant levels of inter-observer variability and sometimes uncertainty in diagnosis. Advances in next-generation sequencing have ushered in new molecular diagnostic frameworks that can improve accuracy, critical in deciding treatment. Methods: We utilized machine learning (XGBoost) and developed comprehensive molecular classifiers for cancer site of origin (22 classes i.e. breast, prostate, lung, etc.) and cancer lineage (8 classes i.e. adenocarcinoma, squamous cell carcinoma, etc.) to augment traditional histopathologic assessment. These models were trained on a total of 8,249 tumor samples. We then evaluated performance using a large independent validation cohort consisting of 10,376 samples. While 8,886 of these were primary tumors from 97 different datasets, uniquely we also assessed performance in 1,490 metastatic tumors from 17 datasets. Pathologic diagnosis of metastatic tumors can be more difficult due to de-differentiation, but metastatic samples for molecular profiling are difficult to obtain and were only rarely included in previous efforts. Results: After model training, we locked the site of origin and lineage expression-based models and next evaluated performance on our independent validation cohort. Overall accuracy was 92.5% for cancer site of origin and 97.2% for cancer lineage on validation. Accuracy among primary site samples was higher than for metastatic samples (93.4%, and 86.8% for cancer site of origin, respectively; 97.5% and 95.7% for cancer lineage, respectively). However, accuracy jumped to an astounding 98-99% for both site of origin and lineage in both primary and metastatic samples where the signatures were highly confident, encompassing the majority of cases. Our unique approach in evaluating site of origin and lineage separately allowed us to identify a lineage differentiation score that was associated with small cell lung and neuroendocrine prostate tumors (AUC 0.955 and 0.833, respectively), as well as worse survival across most evaluable tumor types. Conclusions: To our knowledge, this is the largest and most comprehensive validation of platform-independent site of origin/lineage classifiers to date. Our approach can be applied to any existing research or commercial RNA-seq assay, and provides an objective and quantifiable confidence measurement that is correlated with accuracy. This allows for nuanced interpretation of high vs. low confidence predictions, which can complement and potentially even help guide traditional histopathologic assessment in cancer research, clinical trial design, and ultimately clinical practice.

Toward Automated Field Ballast Condition Evaluation: Algorithm Development Using a Vision Transformer Framework

Ballast degradation in the track substructure may cause poor drainage, settlement, and reduced lateral stability that may affect safety, daily operations, and long-term maintenance of a railroad system. Extreme levels of degradation in the ballast may result in service interruptions because of safety concerns. Therefore, field ballast condition evaluation is deemed crucial. Current state-of-the-practice methods for evaluating ballast condition primarily rely on subjective visual inspection, labor-intensive sampling, and laboratory sieve analyses of collected field samples. A network-level condition assessment of ballast and track substructure is commonly performed using ground-penetrating radar. For site-specific and detailed geotechnical analyses, development of a reliable, accurate, and cost-effective technique for ballast condition evaluation is urgently needed. This paper presents an innovative approach to accomplish image-based ballast condition evaluation based on deep learning techniques. A ballast image dataset (library) is established by collecting images from various railroad sites and laboratory setups of ballast piles. A vision transformer-based segmentation framework is implemented and trained with the established dataset and employed to serve as the image segmentation kernel to relate the image-based Percent Degraded Segments (PDS) with the ground-truth Fouling Index (FI). Based on the presented research findings, the proposed approach for field ballast condition evaluation will serve as the core data analysis component of an automated ballast scanning vehicle to conduct field ballast inspection, which is being developed to serve as an efficient and reliable system for the evaluation of ballast condition in ballasted railroad tracks.

A new adaptive window‐based guided filtering and interpolation for polarization image demosaicing

AbstractThis paper presents a new adaptive window‐based guided filtering and interpolation for polarization image demosaicing. Considering the correlation between the polarization channels, the authors use the total intensity S0 as a guide image to guide the interpolation of all channels. The guided filtering is performed with a new adaptive window size selection technique. A novel iterative residual interpolation is proposed to construct accurate full resolution polarization. Experiments are conducted with the sample images from the demosaicking algorithms dataset. Objective evaluations demonstrate that the proposed algorithm can reduce root mean squared error by at least 27% over other methods. Subjective visual inspection can also verify the improvements in the clarity of objects in angle and degree of the linear polarization images.

Synthetic Data and Computer-Vision-Based Automated Quality Inspection System for Reused Scaffolding

Regular scaffolding quality inspection is an essential part of construction safety. However, current evaluation methods and quality requirements for temporary structures are based on subjective visual inspection by safety managers. Accordingly, the assessment process and results depend on an inspector’s competence, experience, and human factors, making objective analysis complex. The safety inspections performed by specialized services bring additional costs and increase evaluation times. Therefore, a temporary structure quality and safety evaluation system based on experts’ experience and independent of the human factor is the relevant solution in intelligent construction. This study aimed to present a quality evaluation system prototype for scaffolding parts based on computer vision. The main steps of the proposed system development are preparing a dataset, designing a neural network (NN) model, and training and evaluating the model. Since traditional methods of preparing a dataset are very laborious and time-consuming, this work used mixed real and synthetic datasets modeled in Blender. Further, the resulting datasets were processed using artificial intelligence algorithms to obtain information about defect type, size, and location. Finally, the tested parts’ quality classes were calculated based on the obtained defect values.

RECOLORING ISHIMARA TEST IMAGE FOR PROTAN CVD PERSONS

In this paper, the effect of the protan CVD algorithm is analyzed. Using the protan CVD algorithm, theimages are recolored, so that they are suitable for use by protan CVD people. In the first part, a trichromaticanomaly of the protanomaly type is described. After that, the protan CVD algorithm is described. The second part ofthe paper describes the Experiment in which the effectiveness of the CVD algorithm was tested. The CVD algorithmwas applied to images from the Ishihara color blindness test database, which were purposely formed to detect protanCVD persons. The results of the Experiment are shown using images and numerical values of the average chromaticdifference between every two pixels of the image. Finally, a comparative analysis of the results using objective(analysis of chromatic differences) and subjective (visual inspection of the image) methods is presented. Based onthe analysis, an evaluation of the efficiency of the protan CVD algorithm was given.

Real-time assessment of liver fat content using a filter-based Raman system operating under ambient light through lock-in amplification.

During liver procurement, surgeons mostly rely on their subjective visual inspection of the liver to assess the degree of fatty infiltration, for which misclassification is common. We developed a Raman system, which consists of a 1064 nm laser, a handheld probe, optical filters, photodiodes, and a lock-in amplifier for real-time assessment of liver fat contents. The system performs consistently in normal and strong ambient light, and the excitation incident light penetrates at least 1 mm into duck fat phantoms and duck liver samples. The signal intensity is linearly correlated with MRI-calibrated fat contents of the phantoms and the liver samples.

Deep learning-based damage estimation in concrete structures using acoustic emission data

Conventional structural testing methods assess damaged condition of the structures and they generally base on subjective visual inspection. To take necessary precautions before/during a disaster, real-time monitoring of damage progression is very important especially for critical structures such as bridges, viaducts, tunnels, dams and nuclear power plants. Acoustic emission (AE) has been proved to be a successful method for this purpose and it is used to locate even invisible micro-level cracks in the structures. The most important feature that makes this technique advantageous over other non-destructive testing methods is that it continuously monitors damages in structures subjected to load by sensors converting the waves emitted from the damage into electrical signals. By analyzing the parameters of the recorded signals, critical information such as the type, location, origination-time, size and direction of the damage can be obtained. In this study, it was aimed to estimate damage status of the concrete members exposed to loading by developing a deep learning-based damage detection model. This model was developed by monitoring AE activities of concrete and reinforced concrete specimens with different strength, size, properties and loading conditions. Afterwards, a relationship was established between the properties and load levels of the specimens with features of the AE signals. The developed model was tested on reinforced concrete elements under load. The results show that the model is successful at estimating the damage level.