Automated Image Processing Method Research Articles

Prognostic significance of collagen signatures in pancreatic ductal adenocarcinoma obtained from second-harmonic generation imaging

BackgroundPancreatic ductal adenocarcinoma (PDAC) ranks among the deadliest types of cancer, and it will be meaningful to search for new biomarkers with prognostic value to help clinicians tailor therapeutic strategies.MethodsHere we tried to use an advanced optical imaging technique, multiphoton microscopy (MPM) combining second-harmonic generation (SHG) and two-photon excited fluorescence (TPEF) imaging, for the label-free detection of PDAC tissues from a cohort of 149 patients. An automated image processing method was used to extract collagen features from SHG images and the Kaplan-Meier survival analysis and Cox proportional hazards regression were used to assess the prognostic value of collagen signatures.ResultsSHG images clearly show the different characteristics of collagen fibers in tumor microenvironment. We gained eight collagen morphological features, and a Feature-score was derived for each patient by the combination of these features using ridge regression. Statistical analyses reveal that Feature-score is an independent factor, and can predict the overall survival of PDAC patients as well as provide well risk stratification.ConclusionsSHG imaging technique can potentially be a tool for the accurate diagnosis of PDAC, and this optical biomarker (Feature-score) may help clinicians make more approximate treatment decisions.

Quantification of fibrosis extend and airspace availability in lung: A semi-automatic ImageJ/Fiji toolbox.

The evaluation of the structural integrity of mechanically dynamic organs such as lungs is critical for the diagnosis of numerous pathologies and the development of therapies. This task is classically performed by histology experts in a qualitative or semi-quantitative manner. Automatic digital image processing methods appeared in the last decades, and although immensely powerful, tools are highly specialized and lack the versatility required in various experimental designs. Here, a set of scripts for the image processing software ImageJ/Fiji to easily quantify fibrosis extend and alveolar airspace availability in Sirius Red or Masson's trichrome stained samples is presented. The toolbox consists in thirteen modules: sample detection, particles filtration (automatic and manual), border definition, air ducts identification, air ducts walls definition, parenchyma extraction, MT-staining specific pre-processing, fibrosis detection, fibrosis particles filtration, airspace detection, and visualizations (tissue only or tissue and airspace). While the process is largely automated, critical parameters are accessible to the user for increased adaptability. The modularity of the protocol allows for its adjustment to alternative experimental settings. Fibrosis and airspace can be combined as an evaluation of the structural integrity of the organ. All settings and intermediate states are saved to ensure reproducibility. These new analysis scripts allow for a rapid quantification of fibrosis and airspace in a large variety of experimental settings.

Quantitative 2D fruit shape analysis of a wide range of pear genetic resources toward shape design breeding

The shape of a fruit is one of the main characteristics of fruit morphology, and its quantitative description is crucial to improve fruit quality. However, the shape of a pear fruit is relatively complex, and difficult to describe comprehensively. This study aimed to quantitatively analyze the morphological diversity of 2D fruit shapes of 276 cultivars, covering 6 groups including Pyrus pyrifolia, Pyrus bretschneideri, Pyrus ussuriensis, Pyrus sinkiangensis, Pyrus communis and interspecific hybridization, and discover shape traits with high heritability. Through an automated image processing method, the fruit contour was first accurately extracted, and then geometrical parameters and Fourier descriptors of the pear fruit shape were obtained. Furthermore, principal components (PCs) of the Fourier descriptors were analyzed, and the broad-sense heritability (H2) of the PCs and standardized width (WSD) were estimated. After that, the pear shape stability was calculated, and cultivars were clustered by their shape characteristics. This paper mainly discusses the longitudinal section, as the cross-section variation among cultivars is small. For the longitudinal section, the results showed more than 95 % of the two-dimensional shape variation can be explained by only six principal components; WSD20, WSD30 and PC1 have high H2; the shape stability of western pear was the lowest; a new class of shape was found by clustering the shape traits. The results indicated that the proposed pipeline can provide an objective and comprehensive quantitative description of pear fruit shape for a wide range of cultivars, which is necessary to discover the key genes determining pear shape. This approach is also beneficial for research on pear breeding and germplasm resource management.

Focus enhancement in long schlieren imaging systems using corrector lenses.

Long schlieren imaging systems, where the parallel light test section is longer than the focal length of the focusing schlieren optics, have a limited region in the test section in which any occluding object, such as a wind tunnel model, is in focus in the schlieren image. Corrector lenses are introduced here to alter the location in the test section where image focus is achieved. Corrector lenses with focal lengths varying from -1000 to -100m m were studied. Lens-type inline and z-type mirror schlieren systems were experimentally tested with multiple collecting optic focal lengths to characterize the changes in focal position. An automated image processing method was used to determine the plane of best focus from image sequences. The introduction of the corrector lens was observed to move the location of best focus within the schlieren system test section while also causing a decrease in the focal sharpness of the images and altering the magnification. The thin lens equation was found to provide a good estimate of the focal location change in the schlieren imaging systems with the addition of the corrector lens.

Efficacy comparison of two methods for determining the position of the rebate edge (formed after MDF machining) during automatic monitoring of workpiece delamination

: Efficacy comparison of two methods for determining the position of the rebate edge (formed after machining) during automatic monitoring of workpiece delamination. Delamination is one of the most common defects in the processing of wood-based materials. It has a huge impact on the quality of the final product. In order to determine the delamination indicators in a simple and reliable way, the automatic image processing method can be used (Śmietańska et al. 2020). Bator and Śmietańska (2019) proposed the special algorithm to estimate the straight line representing a milling edge. However, this algorithm is quite complicated. The aim of this article is to check whether the aforementioned (complicated) algorithmic way can be replaced by a much simpler idea – the precise manual positioning of the scanned sample on the scanner (using very simple device installed on the scanner). The special experimental research was carried out to compare the effectiveness of the two different methods. The straight line which represents the rebate edge identified by Bator and Śmietańska (2019) algorithm was usually accurate to 1 pixel (0.02 mm). The analogue line based on the assumption that the scanned samples were perfectly positioned on the scanner only sometimes fit just as well. At worst, the distance between these lines is 0.2 mm. Usually the distance did not exceed 0.16 mm but was significant and quite random. There was no statistically significant correlation between this parameter (Dmax) and tool condition (VB). It means that sample were not perfect positioned. They were placed more or less in the same position because of imperfect stiffness of the frame installed on the scanner and human errors.

Detection of fibrotic changes in the progression of liver diseases by label-free multiphoton imaging.

Collagen fibers play an important role in the progression of liver diseases. The formation and progression of liver fibrosis is a dynamic pathological process accompanied by morphological changes in collagen fibers. In this study, we used multiphoton microscopy for label-free imaging of liver tissues, allowing direct detection of various components including collagen fibers, tumors, blood vessels, and lymphocytes. Then, we developed a deep learning classification model to automatically identify tumor regions, and the accuracy reaches 0.998. We introduced an automated image processing method to extract eight collagen morphological features from various stages of liver diseases. Statistical analysis showed significant differences between them, indicating the potential use of these quantitative features for monitoring fibrotic changes during the progression of liver diseases. Therefore, multiphoton imaging combined with automatic image processing method would hold a promising future in rapid and label-free diagnosis of liver diseases.

Detection and mapping of artillery craters with very high spatial resolution satellite imagery and deep learning

Unexploded munitions are some of the most enduring remnants of conflicts around the world. Their effects on the economy, health, environment, and post-conflict rehabilitation are long reaching and devastating for the areas they plague. With the advancements in very high spatial resolution (VHR) satellite multispectral imaging at sub-meter resolution, it becomes possible to detect object attributes at the scale of individual impacts (craters) of heavy weapon shelling. Manual identification and delineation of artillery craters in satellite imagery is time and resource consuming, especially when large territories and volumes of VHR data are considered. Therefore, automatic image processing methods should be explored. Here, we evaluate the application of a deep learning approach for identifying and mapping artillery craters in agricultural fields in Eastern Ukraine during the onset of armed conflict in 2014. The model was applied to pansharpened multispectral VHR imagery acquired by the WorldView-2 satellite at 0.5-m spatial resolution. The model can detect artillery craters with producer's accuracy (PA) (or recall) of 0.671 and user's accuracy (UA) (or precision) of 0.392 in terms of crater area and shape, and PA of 0.559 and UA of 0.427 in terms of binary crater identification. The model's performance is dependent on crater size. Reliability of crater detection and mapping improves as the size of craters increases. For example, for craters larger than 60 m2 PA is 0.803 and UA is 0.449 (per-pixel), and PA is 0.891 and UA is 0.721 (per-object). Overall, the model prioritizes PA over UA, i.e., omission error over commission error, and is better at detecting craters than their shapes. We applied the trained model to a separate, 858 km2 subregion of Donetsk oblast to automatically estimate and map the locations, number and area of artillery craters. Our estimates revealed over 22,000 craters in the subregion, which occupy an area of 1.2 km2, or 0.14% of the region, primarily in agricultural fields. The availability of such crater maps is extremely valuable within demining and chemical decontamination efforts and can assist in assessing the impact of warfare on agriculture and the environment. We outline the current limitations of the proposed approach and avenues for further research for improving artillery crater detection and mapping.

An Automatic Image Processing Method Based on Artificial Intelligence for Locating the Key Boundary Points in the Central Serous Chorioretinopathy Lesion Area.

Accurately and rapidly measuring the diameter of central serous chorioretinopathy (CSCR) lesion area is the key to judge the severity of CSCR and evaluate the efficacy of the corresponding treatments. Currently, the manual measurement scheme based on a single or a small number of optical coherence tomography (OCT) B-scan images encounters the dilemma of incredibility. Although manually measuring the diameters of all OCT B-scan images of a single patient can alleviate the previous issue, the situation of inefficiency will thus arise. Additionally, manual operation is subject to subjective factors of ophthalmologists, resulting in unrepeatable measurement results. Therefore, an automatic image processing method (i.e., a joint framework) based on artificial intelligence (AI) is innovatively proposed for locating the key boundary points of CSCR lesion area to assist the diameter measurement. Firstly, the initial location module (ILM) benefiting from multitask learning is properly adjusted and tentatively achieves the preliminary location of key boundary points. Secondly, the location task is formulated as a Markov decision process, aiming at further improving the location accuracy by utilizing the single agent reinforcement learning module (SARLM). Finally, the joint framework based on the ILM and SARLM is skillfully established, in which ILM provides an initial starting point for SARLM to narrow the active region of agent, and SARLM makes up for the defect of low generalization of ILM by virtue of the independent exploration ability of agent. Experiments reveal the AI-based method which joins the multitask learning, and single agent reinforcement learning paradigms enable agents to work in local region, alleviating the time-consuming problem of SARLM, performing location task in a global scope, and improving the location accuracy of ILM, thus reflecting its effectiveness and clinical application value in the task of rapidly and accurately measuring the diameter of CSCR lesions.

Efficacy comparison of two methods for determining the position of the rebate edge (formed after MDF machining) during automatic monitoring of workpiece delamination

Delamination is one of the most common defects in the processing of wood-based materials. It has a massive impact on the quality of the final product. In order to determine the delamination indicators simply and reliably, the automatic image processing method can be used (Śmietańska et al., 2020). Bator and Śmietańska (2019) proposed a special algorithm to estimate the straight line representing a milling edge. However, this algorithm is quite complicated. The aim of this article is to check whether the aforementioned (complicated) algorithmic way can be replaced by a much simpler idea – the precise manual positioning of the scanned sample on a scanner (using a very simple device installed on the scanner). Special experimental research was carried out to compare the effectiveness of the two different methods. The straight line representing the rebate edge identified by the Bator and Śmietańska (2019) algorithm was usually accurate to 1 pixel (0.02 mm). Based on the assumption that the scanned samples were perfectly positioned on the scanner, the analogue line only sometimes fit just as well. At worst, the distance between these lines is 0.2 mm. Usually, the distance did not exceed 0.16 mm but it was significant and quite random. There was no statistically significant correlation between this parameter (Dmax) and tool condition (VB). It means that the samples were not perfectly positioned. They were placed, more or less, in the same position because of the imperfect stiffness of the frame installed on the scanner and human errors.

Weighted voting ensemble method for predicting workpiece imaging dimensional deviation based on monocular vision systems

Due to the different heights and positions of all kinds of workpieces, the imaging dimension of workpieces in monocular vision systems exists in deviation. To reduce the deviation and improve the accuracy of dimensional measurement, a BRR-SVR-BPNN weighted voting ensemble learning deviation prediction algorithm is proposed. The proposed method combines the probability inference feature of Bayesian Ridge regression (BRR), the maximum interval penalty feature of support vector regression (SVR), and the nonlinear expression feature of the backpropagation neural network (BPNN). And an automatic image processing method based on Halcon is proposed to establish the datasets for the deviation prediction algorithm. The experimental results show that the proposed algorithm has well prediction effect, high robustness, and accuracy on deviation prediction. It can be easily applied to predict the imaging dimension deviation of various workpieces or other similar tasks to improve the accuracy of workpiece dimension measurement in the industrial field.

How innovations in methodology offer new prospects for volume electron microscopy

Detailed knowledge of biological structure has been key in understanding biology at several levels of organisation, from organs to cells and proteins. Volume electron microscopy (volume EM) provides high resolution 3D structural information about tissues on the nanometre scale. However, the throughput rate of conventional electron microscopes has limited the volume size and number of samples that can be imaged. Recent improvements in methodology are currently driving a revolution in volume EM, making possible the structural imaging of whole organs and small organisms. In turn, these recent developments in image acquisition have created or stressed bottlenecks in other parts of the pipeline, like sample preparation, image analysis and data management. While the progress in image analysis is stunning due to the advent of automatic segmentation and server‐based annotation tools, several challenges remain. Here we discuss recent trends in volume EM, emerging methods for increasing throughput and implications for sample preparation, image analysis and data management.

Design and Implementation of an Unreal Engine 4-Based Smart Traffic Control System for Smart City Applications

Traffic congestion is a serious problem nowadays, especially in Dhaka city. With the increasing population and automation, it has become one of the most critical issues in our country. There can be a lot of causes of congestion in traffic, such as insufficient capacity, large red signal delay, unrestrained demand, etc, which causes extra time delay, extra fuel consump-tion, a speed reduction of vehicle, and financial loss. The traffic control system is one of the most important factors affecting traffic flow. Poor traffic management around these hotspots could result in prolonged traffic jams. Small critical locations that are frequent hotspots for congestion are a common byproduct of poorly constructed road networks in many developing countries. In this research, we first offer a straightforward automated image processing method for analyzing CCTV camera image feeds to determine the level of traffic congestion. Our system’s design seeks to use real-time photos from the cams at traffic intersections to calculate traffic density using image processing, and to adjust the traffic signal on the data obtained based on the current traffic congestion on the road. We suggest tailoring our system to erratic traffic feeds with poor visual quality. Using live Surveillance camera feeds from multiple traffic signals in Dhaka city, we demonstrate evidence of this bottleneck breakdown tendency persisting over prolonged time frames across multiple locations. To partially address this problem, we offer a local adapting algorithm that coordinates signal timing behavior in a restricted area and can locally minimize congestion collapse by maintaining time-variant traffic surges. Using simulation-based research on basic network topologies, we show how our local decongestion protocol may boost the capacity of the road and avoid traffic collapse in limited scenarios.

How Many Bedrooms Do You Need? A Real-Estate Recommender System from Architectural Floor Plan Images

This paper introduces an automated image processing method to analyze an architectural floor plan database. The floor plan information, such as the measurement of the rooms, dimension lines, and even the location of each room, can be automatically produced. This assists the real-estate agents to maximise the chances of the closure of deals by providing explicit insights to the prospective purchasers. With a clear idea about the layout of the place, customers can quickly make an analytical decision. Besides, it reduces the specialized training cost and increases the efficiency in business actions by understanding the property types with the greatest demand. Succinctly, this paper utilizes both the traditional image processing and convolutional neural networks (CNNs) to detect the bedrooms by undergoing the segmentation and classification processes. A thorough experiment, analysis, and evaluation had been performed to verify the effectiveness of the proposed framework. As a result, a three-class bedroom classification accuracy of ∼ 90% was achieved when validating on more than 500 image samples that consist of the different room numbers. In addition, qualitative findings were presented to manifest visually the feasibility of the algorithm developed.

Hydrodynamics study of a fast liquid–liquid oxidation process with in situ gas production in microreactors

AbstractHydrodynamics characteristics of a fast and highly exothermic liquid–liquid oxidation process with in situ gas production in microreactors were studied using a newly developed experimental method. In the adipic acid synthesis through the K/A oil (the mixture of cyclohexanol and cyclohexanone) oxidation with nitric acid, bubble generation modes were divided into four categories. The gas production became more intensive, unstable, even explosive with increasing the oil phase feed rate and the temperature. A novel automatic image processing method was developed to monitor the instantaneous velocity online by tracking the gas–liquid interface. The axial velocity at the same location was unstable due to the changing gas production rate. Furthermore, the actual residence time was obtained easily with being only 36% of the space–time minimally, beneficial for establishing accurate kinetics and mass transfer models with time participation. Finally, an empirical correlation was developed to predict the actual residence time under different conditions.

On a hybrid lossless compression technique for three-dimensional medical images.

In the last two decades, incredible progress in various medical imaging modalities and sensing techniques have been made, leading to the proliferation of three‐dimensional (3D) imagery. Byproduct of such great progress is the production of huge volume of medical images and this big data place a burden on automatic image processing methods for diagnostic assistance processes. Moreover, large amount of medical imaging data needs to be transmitted with no loss of information for the purpose of telemedicine, remote diagnosis etc. In this work, we consider a hybrid lossless compression technique with object‐based features for three‐dimensional (3D) medical images. Our approach utilizes two phases as follows: first we determine the volume of interest (VOI) for a given 3D medical imagery using selective bounding volume (SBV) method, and second the obtained VOI is encoded using a hybrid lossless algorithm using Lembel‐Ziv‐Welch Coding (LZW) followed by arithmetic coding (L to A). Experimental results show that our proposed 3D medical image compression method is comparable with other existing standard lossless encoding methods such as Huffman Coding, Run Length Coding, LZW, and Arithmetic Coding and obtains superior results overall.

Assessment of the effectiveness of a rockfall ditch through 3-D probabilistic rockfall simulations and automated image processing

Abstract Rockfall ditches or catchment areas aim to collect falling blocks at the toe of a source zone by dissipating the energy of blocks in an excavated trench. The effectiveness of a rockfall ditch is simply expressed by its block catchment performance and can be evaluated by empirically using existing design charts as well as rockfall simulations. Although 2-dimensional (2-D) analysis has been executed to assess the catchment ditch effectiveness in engineering practice, 3-dimensional (3-D) rockfall models have not received enough attention so far. In this study, the effectiveness of a considerably long rockfall ditch to protect a settlement from falling rocks was assessed on the basis of 3-D rockfall analyses executed using high-resolution digital surface models. The rockfall ditch efficiency was found to be moderate to limited for various segments considering the percentage of blocks not trapped by the ditch. Moreover, the sensitivity of ditch efficiency to ditch depth was analyzed by automated image processing method as well. Additionally, a particular section of ditch alignment was fictitiously excavated or filled by synthetic Digital Surface Model (DSM) generation through image processing. 3-D rockfall modeling carried out using the DSMs with synthetically manipulated ditches points out that the effectiveness of a catchment ditch is highly depended upon ditch depth. Even a small volume of block accumulation inside the ditch definitely reduces the ditch performance resulting extended runout distances reaching to residential area. Finally, 3-D rockfall modeling is accepted to be an effective tool to rate the efficiency of existing rockfall ditches and synthetically generated ditches on DSMs (or DEMs) by means of automated image processing method may assist the control of current ditch dimension as well as new catchment ditch design.

Soot aggregate morphology deduced from thermophoretic sampling in coflow laminar methane diffusion flames at pressures up to 30 bar

A high pressure combustion chamber, housing a co-flow burner and a novel thermophoretic soot sampling system, was used to collect soot samples from methane-air diffusion flames at pressures from 5 bar to 30 bar. To have tractable diffusion flames with the same residence times, constant methane flow rate of 0.55 mg/s was maintained throughout the experiments. Soot aggregates were collected on 3 mm diameter transmission electron microscope (TEM) grids and the images of the soot aggregates were captured by a TEM device. Sampling was done at three heights above the burner rim at 3, 6, and 8 mm. An automated image processing method was used to analyze the TEM images of soot aggregates to infer the primary soot particle diameters and the aggregate morphology. The mean primary soot particle diameters increased from 5 to 15 bar at all three sampling locations and seemed to reach a peak around 15–20 bar at 6 and 8 mm height and start decreasing with further increase in pressure. On the other hand, mean primary soot diameter at 3 mm height increased slightly beyond 15 bar. In view of the measured soot concentrations and temperatures at the sampling locations at 3, 6, and 8 mm above the burner, the soot number density is argued to be increasing with increasing pressure. Throughout the all experimental conditions, fractal dimension of the aggregates in these methane diffusion flames did not exceed 1.4 which is much smaller than the accepted universal value of about 1.85. Morphological characteristics inferred from TEM images indicate that the aggregation is less prominent in methane flames in comparison to ethylene diffusion flames under similar conditions.

Assessment of two automated image processing methods to estimate bubble size in industrial flotation machines

This Short Communication presents a comparison between two automated methodologies to estimate bubble size in industrial flotation equipment (mechanical cells and columns). The studied database includes 106 conditions in flotation machines from 10 to 300 m3, which were sampled by means of the McGill bubble size analyser. The first methodology (conventional image processing) uses circularity, ellipse detection, segmentation and reprocessing of separated objects to estimate the Sauter mean diameter D32. The second methodology evaluates the binary images as trains of pulses, whose spectral bandwidth is correlated with the D32. Both techniques have been compared to a semi-automatic approach, which manually complete the conventional image analysis to obtain the total bubble size distribution. For D32 < 2 mm (spherical regime), the first approach shows better performance than the spectral analysis, with significantly higher precision. For D32 ≥ 2 mm (ellipsoidal regime), the conventional image processing presents significant bias, high dispersion as well as a cliff trend for D32 > 3.5 mm. On the other hand, the spectral method showed a better sensitivity with an approximately linear trend up to 5 mm. Thus, a combination of these two methodologies is suggested for the D32 characterization in industrial flotation machines.

Schiefer counter: An alternative method for clonogenic assay evaluation

IntroductionClonogenic assay evaluates the potential of cells to undergo division or generate clones following treatment with a chemical or other agent, thereby allowing the evaluation of cytotoxic and/or antiproliferative effects. Clonogenic assay analysis using traditional methods tends to be time-consuming and yield inconsistent results, whereas results from analyses conducted using automated image processing methods may be misleading or subject to misinterpretation. Thus, the aim of this work was to validate and demonstrate the applicability of a recently developed software. MethodsRepeatability of measurements was evaluated by comparing results from 10 replicate images from a single well. To evaluate the viability of the software, results were compared with those obtained from manual counting, crystal violet optical density, and up-to-date automated methods. A clonogenic index was experimentally developed using the individual area occupied by colonies, while clone stratification was used to differentiate between antiproliferative and cytotoxic effects. ResultsThe developed software showed to be a reliable and consistent tool for clonogenic assay evaluation, presenting a repeatability mean error of 0.79% for the number of colonies and 0.89% for the total area of colonies, as well as exhibiting a significant correlation (p < 0.05) with results obtained from widely adopted gold standard methods. The software was also able to detect an appropriate dose-dependent effect as well as a predominant cytotoxic effect of vincristine on MCF-7 cells and calculate the clonogenic index. DiscussionTherefore, this software is adequate for the analysis of clonogenic assay images, differentiating between cytotoxic and antiproliferative trends.

Multiple sclerosis lesion activity segmentation with attention-guided two-path CNNs

Multiple sclerosis is an inflammatory autoimmune demyelinating disease that is characterized by lesions in the central nervous system. Typically, magnetic resonance imaging (MRI) is used for tracking disease progression. Automatic image processing methods can be used to segment lesions and derive quantitative lesion parameters. So far, methods have focused on lesion segmentation for individual MRI scans. However, for monitoring disease progression, lesion activity in terms of new and enlarging lesions between two time points is a crucial biomarker. For this problem, several classic methods have been proposed, e.g., using difference volumes. Despite their success for single-volume lesion segmentation, deep learning approaches are still rare for lesion activity segmentation. In this work, convolutional neural networks (CNNs) are studied for lesion activity segmentation from two time points. For this task, CNNs are designed and evaluated that combine the information from two points in different ways. In particular, two-path architectures with attention-guided interactions are proposed that enable effective information exchange between the two time point’s processing paths. It is demonstrated that deep learning-based methods outperform classic approaches and it is shown that attention-guided interactions significantly improve performance. Furthermore, the attention modules produce plausible attention maps that have a masking effect that suppresses old, irrelevant lesions. A lesion-wise false positive rate of 26.4% is achieved at a true positive rate of 74.2%, which is not significantly different from the interrater performance.