Manual Error Research Articles

Evolutionary transfer optimization-based approach for automated ictal pattern recognition using brain signals.

The visual scrutinization process for detecting epileptic seizures (ictal patterns) is time-consuming and prone to manual errors, which can have serious consequences, including drug abuse and life-threatening situations. To address these challenges, expert systems for automated detection of ictal patterns have been developed, yet feature engineering remains problematic due to variability within and between subjects. Single-objective optimization approaches yield less reliable results. This study proposes a novel expert system using the non-dominated sorting genetic algorithm (NSGA)-II to detect ictal patterns in brain signals. Employing an evolutionary multi-objective optimization (EMO) approach, the classifier minimizes both the number of features and the error rate simultaneously. Input features include statistical features derived from phase space transformations, singular values, and energy values of time-frequency domain wavelet packet transform coefficients. Through evolutionary transfer optimization (ETO), the optimal feature set is determined from training datasets and passed through a generalized regression neural network (GRNN) model for pattern detection of testing datasets. The results demonstrate high accuracy with minimal computation time (<0.5 s), and EMO reduces the feature set matrix by more than half, suggesting reliability for clinical applications. In conclusion, the proposed model offers promising advancements in automating ictal pattern recognition in EEG data, with potential implications for improving epilepsy diagnosis and treatment. Further research is warranted to validate its performance across diverse datasets and investigate potential limitations.

Current Trends of Artificial Intelligence in Biotechnology

Artificial intelligence (AI) is already widely used in biotechnology to solve a variety of problems. These include, for example, drug discovery, drug safety, functional and structural proteomics/genomics, metabolomics, pharmacology, pharmacogenetics and pharmacogenomics, among many others. Future advances in this domain depend critically on the ability of biotechnology researchers to use advanced AI solutions effectively. The biotechnology industry currently relies heavily on data storage, filtering, analysis and sharing. Biotechnology companies and various healthcare organizations around the world already maintain huge data bases. Drug manufacturing, chemical analysis of various compounds, sequencing of RNA and DNA, enzyme studies, and other similar biological processes all require strong support from AI software solution to move faster and reduce manual errors. It is important to emphasize at the very beginning that all the successful AI we are describing today relies entirely on digital technology to function. Digitalization is therefore the very first step towards any AI application. In many cases, AI systems are integrated with other digital technologies such as sensors, actors (cyber-physical systems (CPS), often just called robots), and technology to enable the automation of tasks and the collection and analysis of data. Overall, the development and use of AI is dependent on digital technology - the basis for it is digital computers. Digital transformation refers to the use of digital technologies to fundamentally change the way companies, organizations, research institutions and universities operate. In the context of biotechnology, digital transformation can involve the introduction of new technologies and processes to improve the efficiency, accuracy, and speed of research and development and enable the development of entirely new and disruptive products and services. Digital transformation can help accelerate the development and use of AI in biotechnology by providing access to big data and automating certain tasks, which can help improve the efficiency and accuracy of research and development. In this Editorial, it has been clearly stated what exactly AI means, concomitant with explaining the specific differences between AI, machine learning, and deep learning to provide a worthy common understanding. Afterwards, there is successful introduction of significant domains of biotechnology where AI is being applied or may be applicable in the future. Thereafter, this editorial presents certain cross-cutting challenges where it is predominantly significant to advance future studies

Transforming Echocardiography: The Role of Artificial Intelligence in Enhancing Diagnostic Accuracy and Accessibility.

Artificial intelligence (AI) has shown transformative potential in various medical fields, including diagnostic imaging. Recent advances in AI-driven technologies have opened new avenues for improving echocardiographic practices. AI algorithms enhance the image quality, automate measurements, and assist in the diagnosis of cardiovascular diseases. These technologies reduce manual errors, increase consistency, and match the diagnostic performances of experienced echocardiographers. AI in tele-echocardiography offers significant benefits, particularly in rural and remote regions in Japan, where healthcare provider shortages and geographic isolation hinder access to advanced medical care. AI enhances accessibility, provides real-time remote analyses, supports continuous monitoring, and improves the quality and efficiency of remotely delivered cardiac care. However, addressing challenges related to data security, transparency, integration into clinical workflows, and ethical considerations is essential for the successful implementation of AI in echocardiography. On overcoming these challenges, AI will be able to revolutionize echocardiography and ensure timely and effective cardiac care for all patients in the future.

The role of e-procurement for sustainable supply chain management: The case of Malawi Revenue Authority (MRA)

This study has assessed the potential of e-procurement in improving the sustainability of the MRA's supply chain. According to the results of the study, e-procurement can enhance transparency by 70%, reduce manual errors by 50%, and make staff more efficient by 60%. These improvements therefore allow better monitoring of supplier practices and increased attention to strategic activities, which involve greater consideration of sustainability issues within the practices concerned. It is further envisioned to increase stakeholder engagement by 65%, which will yield improved information and sustainability-driven procurement decisions. A cost-benefit analysis indicates a likely cost reduction of 40% in five years, thus further consolidating the financial case for e-procurement. From these observations, it seems that MRA's shift to e-procurement is likely to successfully enhance supply chain sustainability so that procurement practices are oriented towards the set national goals of sustainability and contribute towards sustainable development in Malawi.

A digital taxation model for Nigeria: standardizing collection through technology integration

This paper introduces a digital taxation model for Nigeria, aimed at standardizing tax collection through advanced technology integration. The model addresses the critical need for modernizing Nigeria’s tax administration to enhance efficiency, accuracy, and compliance in tax collection processes. By leveraging digital technologies, the proposed model seeks to streamline tax administration, reduce manual errors, and improve overall tax revenue management. The model advocates for the implementation of a comprehensive digital tax infrastructure, including the development of an integrated tax management system, digital payment platforms, and real-time data analytics tools. This approach emphasizes the use of application programming interfaces (APIs) for seamless interaction between tax authorities, financial institutions, and taxpayers, ensuring a unified and efficient tax collection system. Blockchain technology is proposed to enhance transparency and traceability, providing an immutable ledger for all transactions, which helps in mitigating fraud and increasing taxpayer confidence. The digital taxation model also includes automated compliance monitoring and reporting systems that utilize artificial intelligence (AI) for predictive analytics and risk assessment. These systems are designed to identify discrepancies, flag potential non-compliance issues, and provide actionable insights to both tax authorities and businesses. Additionally, the model emphasizes the importance of taxpayer education and support, advocating for user-friendly digital platforms and resources to facilitate smooth adoption and compliance. Expected outcomes of the model include increased tax revenue through improved compliance and reduced tax evasion, enhanced operational efficiency for tax authorities, and a more transparent and equitable taxation process. By aligning with global best practices and integrating cutting-edge technologies, the digital taxation model aims to transform Nigeria’s tax administration landscape, fostering a more robust and reliable financial system.

Case Study of Inventory Management for Biology Laboratory Materials using ERP System

Efficient inventory management is essential for biology laboratories to ensure the availability of materials for research and experiments. This case study explores the implementation of an Enterprise Resource Planning (ERP) system for inventory management in a biology laboratory. The objective is to enhance inventory control, streamline processes, and improve overall efficiency. The ERP system was implemented to address inventory management challenges and improve the accuracy and transparency of material tracking. The study assessed the impact of the ERP system on inventory control, procurement processes, and decision-making related to material planning. The findings of the case study demonstrate the benefits of implementing an ERP system for inventory management in a biology laboratory. The system provided real-time visibility of stock levels, enabling efficient tracking and monitoring of laboratory materials. It also automated procurement processes, reducing manual errors and streamlining the procurement cycle. The ERP system facilitated informed decision-making through data analytics, aiding in material planning and resource allocation. The successful implementation of the ERP system in the biology laboratory serves as a model for other similar laboratories seeking to enhance their inventory management practices. The case study emphasizes the significance of leveraging technology, such as ERP systems, to optimize inventory control and improve operational efficiency. The benefits include improved accuracy, reduced wastage, and increased productivity. This case study contributes to the existing knowledge on inventory management for biology laboratory materials and provides practical insights for organizations considering the implementation of an ERP system.

Analysis of ChatGPT's capabilities for solving problems of reverse-engineering of software

The article presents the results of the analysis and evaluation of the capabilities of the open version of the chatbot with artificial intelligence ChatGPT 3.5 for solving typical problems of software reverse engineering. Three classes of reverse engineering tasks were selected for analysis: source code analysis, binary code analysis, and data models analysis. In each class of tasks, the most typical tasks were selected taking into account the limitations of ChatGPT regarding the processing of graphical models and the amount of input and output data, and sets of test tasks were developed for each task. As an approach to assessment, an approach similar to the assessment of competencies of higher education applicants after studying the relevant discipline was chosen. The following were considered as criteria for evaluating answers to test tasks: correctness (rightness, coincidence with expectations); completeness (obtaining the final result); accuracy (solving the task without additional questions); reasonableness (availability of explanations and answers to questions). The following scale was used and described for evaluations: excellent, very good, good, satisfactory, sufficient, unsatisfactory. During the testing for each test task, tasks statement and all necessary data were entered through the ChatGPT interface. As a result of the analysis, it was determined that ChatGPT better solves the problems of source code analysis (excellent and very good grades were obtained for semantic and structural analysis, restoration of mathematical support, quality assessment, security audit and refactoring, as well as for conversion to another programming language), tasks of decompilation of IDA pseudocode into complete C source code, tasks of reverse engineering of relational databases, and tasks of generating YARA rules for recognizing file formats. Unsatisfactory grades were obtained for dynamic analysis of assembly code and determination of binary file structures of non-standard formats. ChatGPT solves the rest of the problems well and satisfactorily, but requires checking the results, clarifying queries and prompts, as well as manual error correction in some cases. ChatGPT errors were observed when analyzing binary data represented by hexadecimal symbols, as well as errors in generated scripts for programming in IDA. On the basis of the set grades, conclusions were made regarding the expediency, possibility, or impracticality of using ChatGPT to solve each type of software reverse engineering problems, and appropriate recommendations were also provided. Prospects for further research include testing new versions of ChatGPT and other similar artificial intelligence systems regarding the capabilities of analyzing and synthesizing graphical models of software.

Microscopic urinary particle detection by different YOLOv5 models with evolutionary genetic algorithm based hyperparameter optimization

The diagnosis of kidney disease often involves analysing urine sediment particles. Traditionally, urinalysis was performed manually by collecting urine samples and using a centrifuge, which was prone to manual errors and relied on labour-intensive processes. Automated urine sediment microscopy, based on machine learning models, requires segmentation and feature extraction, which can hinder model performance due to intrinsic characteristics of microscopic images. Deep learning models based on convolutional neural networks (CNNs) often rely on a large number of manually annotated data, making the system computationally complex. This study propose an advanced deep learning model based on YOLOv5, which offers faster performance and requires comparatively less data. The proposed model used five variants of the YOLOv5 model (YOLOv5n, YOLOv5s, YOLOv5m, YOLOv5l, and YOLOv5x) to detect six categories of urine particles (erythrocyte, leukocyte, crystals, cast, mycete, epithelial cells) from microscopic urine sediment images. The dataset involved 5376 images of urine sediments with 6 particles. There are 30 sets of hyperparamreteres are employed in the YOLOv5 model. To optimize the hyperparameters and fine-tune with the urine sediment dataset and for training each model, the system employed a genetic algorithm (GA) based on evolutionary principles named as Evolutionary Genetic Algorithm (EGA). Among the six categories of detected particles mycete achieved maximum performance with a mAP of 97.6 % and crystals achieved minimum performance with a mAP of 81.7 % with YOLOv5x model compared to other particles. To optimize the hyperparameters for training each model, the system employed a genetic algorithm (GA) based on evolutionary principles named as Evolutionary Genetic Algorithm (EGA). Among all the models, YOLOv5l and YOLOv5x performed the best. YOLOv5l achieved a mean average precision (mAP) of 85.8 % while YOLOv5x achieved a mAP of 85.4 % at an IoU threshold of 0.5. The detection speed per image was 23.4 ms for YOLOv5l and 28.4 ms for YOLOv5x. The proposed method developed a faster and better automated microscopic model using advanced deep learning techniques to detect urinary particles from microscopic urine sediment images for kidney disease identification. The method demonstrated strong performance in urinalysis.

Design and development of a low-cost device to measure foot parameters in children with disability: A “footmeter”

AIM: To design and develop a low-cost device to measure foot parameters in children with disability. METHOD: Designing of the device was done using Solidworks software, after finalization of design, Phase 2 which involved the development of footmeter, materials used were, a rectangular wooden frame, an aluminum slider, goniometers, metal sheet with markings of 1 cm over an entire sheet in vertical as well as horizontal direction forming small grids. RESULT: An early version of “Footmeter” device was developed by assembling all the parts together. The Footmeter evaluates: foot length, width of forefoot, midfoot and hindfoot, height of Medial Longitudinal Arch(MLA) , abduction of 1st MetaTarso Phalangeal (MTP) and deviation for metatarsus adductus. CONCLUSION: The groundbreaking Footmeter device is meticulously designed for measuring foot parameters in children with disabilities. It offers an unparalleled blend of precision, efficiency, and simplicity. With its objective outcomes, minimal training requirements, decreased manual errors, and a streamlined assessment process, “Footmeter” device stands as the ultimate all-in-one tool for the meticulous measurement of foot parameters in children with disabilities.

A novel methodology for characterizing DFB laser wavelength response (WR) in wavelength modulation spectroscopy (WMS)

A novel methodology to directly characterize the wavelength response (WR) of a DFB laser in wavelength modulation spectroscopy (WMS) is proposed. This method circumvents the cumbersome etalon signal peak counting process which brings in manual errors easily. The particle swarm optimization (PSO) algorithm was employed to effectively address the challenges encountered during the fitting process, which involved as many as 7 parameters in the WR. The WR with different combinations of frequencies of scan and modulation, as well as modulation amplitudes were characterized by this method. And it is shown that there exist negligible differences when comparing the results with those by traditional etalon peaking counting method. In order to verify the accuracy of this method, two tunable diode lasers emitting near 7185.59 cm−1 and 7444.35 cm−1 were employed to perform the measurements of temperature above a burner with calibration-free WMS-2f/1f and direct absorption spectroscopy (DAS), respectively. The temperature measured by two techniques shows great accordance, which indicates the accuracy of WR deduced by the proposed method.

Streamlined Yearbook Automation System

The primary function of online company year book development is to ensure the correct upkeep of personnel records. The ability to articulate one's thoughts and feelings is a great benefit of this medium. Staff members can access all company-related information, including Staff Details and corporate function details, in one convenient location. Both during and after their time as students, administrators keep meticulous records. Access to employee and company details can only be granted by the administrator. After that, students need to get into the development of the Year Book. The website allows for the reduction of manual processes, errors, and time consumption. That's right; computer technology is pervasive across all industries. That is why the invention of computers was crucial in making previously insurmountable jobs much easier. This Online Year Book Development project is designed to keep track of workers' work records and achievements for a specific academic time. This is where the new user has to fill out their details. Once they log in, users will be able to view the company's profile, staff data, and a gallery of the company's work.

Discover, Explain, Improve: An Automatic Slice Detection Benchmark for Natural Language Processing

Abstract Pretrained natural language processing (NLP) models have achieved high overall performance, but they still make systematic errors. Instead of manual error analysis, research on slice detection models (SDMs), which automatically identify underperforming groups of datapoints, has caught escalated attention in Computer Vision for both understanding model behaviors and providing insights for future model training and designing. However, little research on SDMs and quantitative evaluation of their effectiveness have been conducted on NLP tasks. Our paper fills the gap by proposing a benchmark named “Discover, Explain, Improve (DEIm)” for classification NLP tasks along with a new SDM Edisa. Edisa discovers coherent and underperforming groups of datapoints; DEIm then unites them under human-understandable concepts and provides comprehensive evaluation tasks and corresponding quantitative metrics. The evaluation in DEIm shows that Edisa can accurately select error-prone datapoints with informative semantic features that summarize error patterns. Detecting difficult datapoints directly boosts model performance without tuning any original model parameters, showing that discovered slices are actionable for users.1

Positioning errors of anatomical landmarks identified by fixed vertices in homologous meshes

BackgroundHuman movement analysis is usually achieved by tracking markers attached to anatomical landmarks with photogrammetry. Such marker-based systems have disadvantages that have led to the development of markerless procedures, although their accuracy is not usually comparable to that of manual palpation procedures. New motion acquisition systems, such as 3D temporal scanners, provide homologous meshes that can be exploited for this purpose. Research questionCan fixed vertices of a homologous mesh be used to identify anatomical landmarks with an accuracy equivalent to that of manual palpation? MethodsWe used 3165 human shape scans from the CAESAR dataset, with labelled locations of anatomical landmarks. First, we fitted a template mesh to the scans, and assigned a vertex of that mesh to 53 anatomical landmarks in all subjects. Then we defined a nominal vertex for each landmark, as the more centred vertex out of the set assigned for that landmark. We calculated the errors of the template-fitting and the nominal vertex determination procedures, and analysed their relationship to subject’s sex, height and body mass index, as well as their size compared to manual palpation errors. ResultsThe template-fitting errors were below 5 mm, and the nominal vertex determination errors reached maximum values of 24 mm. Except for the trochanter, those errors were the same order of magnitude or smaller than inter-examiner errors of lower limb landmarks. Errors increased with height and body mass index, and were smaller for men than for women of the same height and body mass index. SignificanceWe defined a set of vertices for 53 anatomical landmarks in a homologous mesh, which yields location errors comparable to those obtained by manual palpation for the majority of landmarks. We also quantified how the subject’s sex and anthropometric features can affect the size of those errors.

Setup Uncertainty of Pediatric Brain Tumor Patients Receiving Proton Therapy: A Prospective Study.

This study quantifies setup uncertainty in brain tumor patients who received image-guided proton therapy. Patients analyzed include 165 children, adolescents, and young adults (median age at radiotherapy: 9 years (range: 10 months to 24 years); 80 anesthetized and 85 awake) enrolled in a single-institution prospective study from 2020 to 2023. Cone-beam computed tomography (CBCT) was performed daily to calculate and correct manual setup errors, once per course after setup correction to measure residual errors, and weekly after treatments to assess intrafractional motion. Orthogonal radiographs were acquired consecutively with CBCT for paired comparisons of 40 patients. Translational and rotational errors were converted from 6 degrees of freedom to a scalar by a statistical approach that considers the distance from the target to the isocenter. The 95th percentile of setup uncertainty was reduced by daily CBCT from 10 mm (manual positioning) to 1-1.5 mm (after correction) and increased to 2 mm by the end of fractional treatment. A larger variation existed between the roll corrections reported by radiographs vs. CBCT than for pitch and yaw, while there was no statistically significant difference in translational variation. A quantile mixed regression model showed that the 95th percentile of intrafractional motion was 0.40 mm lower for anesthetized patients (p=0.0016). Considering additional uncertainty in radiation-imaging isocentricity, the commonly used total plan robustness of 3 mm against positional uncertainty would be appropriate for our study cohort.

Paper Invoice Dilemma and Electronic Invoice Solution

This paper will start from the definition and purpose of paper invoice, analyze the shortcomings and problems in the use of paper invoice, and introduce the definition, advantages and purpose of electronic invoice, in order to solve the dilemma faced by the use of paper invoice. As a traditional form of bill, paper invoice plays an important role in commercial transactions, but there are some problems in its use. First of all, paper invoices are prone to damage, loss or tampering, increasing the risk and uncertainty of transactions. Secondly, the preservation and management of paper invoices requires a lot of time, space and labor costs, which brings no small burden to enterprises. In addition, because the paper invoice cannot realize real-time information sharing and tracking, resulting in the information process is not smooth and inefficient. In order to solve these problems of paper invoice, electronic invoice came into being. Electronic invoice is a new form of bill based on electronic data exchange and digital signature technology. It has the following advantages: First, electronic invoice does not need paper carrier, reduces the cost of use and management, and is not easy to be damaged, lost or tampered with, improving the reliability and security of transactions. Secondly, electronic invoices can realize real-time information sharing and tracking, which improves the efficiency and convenience of transactions. In addition, the electronic invoice can also be seamlessly connected with the enterprise's financial management system, to achieve automated accounting processing, reducing manual operations and errors. In short, there are a series of shortcomings and problems in the use of paper invoices, such as easy loss, difficult management, not easy to share information and so on. As a solution, electronic invoice has been more and more widely used and promoted by virtue of its advantages of eliminating paper carrier, improving transaction security and reliability, and realizing real-time information sharing and tracking. In the future, with the continuous development and improvement of technology, electronic invoices will become the mainstream form of business transactions, and promote the process of paper invoices to electronic transformation.

SAGA: A Scalable Framework for Optimizing Data Cleaning Pipelines for Machine Learning Applications

In the exploratory data science lifecycle, data scientists often spent the majority of their time finding, integrating, validating and cleaning relevant datasets. Despite recent work on data validation, and numerous error detection and correction algorithms, in practice, data cleaning for ML remains largely a manual, unpleasant, and labor-intensive trial and error process, especially in large-scale, distributed computation. The target ML application---such as classification or regression models---can be used as a signal of valuable feedback though, for selecting effective data cleaning strategies. In this paper, we introduce SAGA, a framework for automatically generating the top-K most effective data cleaning pipelines. SAGA adopts ideas from Auto-ML, feature selection, and hyper-parameter tuning. Our framework is extensible for user-provided constraints, new data cleaning primitives, and ML applications; automatically generates hybrid runtime plans of local and distributed operations; and performs pruning by interesting properties (e.g., monotonicity). Instead of full automation---which is rather unrealistic---SAGA simplifies the mechanical aspects of data cleaning. Our experiments show that SAGA yields robust accuracy improvements over state-of-the-art, and good scalability regarding increasing data sizes and number of evaluated pipelines.

PhylteR: Efficient Identification of Outlier Sequences in Phylogenomic Datasets.

In phylogenomics, incongruences between gene trees, resulting from both artifactual and biological reasons, can decrease the signal-to-noise ratio and complicate species tree inference. The amount of data handled today in classical phylogenomic analyses precludes manual error detection and removal. However, a simple and efficient way to automate the identification of outliers from a collection of gene trees is still missing. Here, we present PhylteR, a method that allows rapid and accurate detection of outlier sequences in phylogenomic datasets, i.e. species from individual gene trees that do not follow the general trend. PhylteR relies on DISTATIS, an extension of multidimensional scaling to 3 dimensions to compare multiple distance matrices at once. In PhylteR, these distance matrices extracted from individual gene phylogenies represent evolutionary distances between species according to each gene. On simulated datasets, we show that PhylteR identifies outliers with more sensitivity and precision than a comparable existing method. We also show that PhylteR is not sensitive to ILS-induced incongruences, which is a desirable feature. On a biological dataset of 14,463 genes for 53 species previously assembled for Carnivora phylogenomics, we show (i) that PhylteR identifies as outliers sequences that can be considered as such by other means, and (ii) that the removal of these sequences improves the concordance between the gene trees and the species tree. Thanks to the generation of numerous graphical outputs, PhylteR also allows for the rapid and easy visual characterization of the dataset at hand, thus aiding in the precise identification of errors. PhylteR is distributed as an R package on CRAN and as containerized versions (docker and singularity).

Differentiable sampling based efficient architecture search for automatic fault diagnosis

Intelligent diagnosis on rotating machinery has developed rapidly, but different methods have fluctuating performance and fussy design, causing poor effect in practical applications. Thus, it would be great to automatically generate the optimal method for given diagnosis tasks, as differentiable neural architecture search (DNAS) does. However, three challenges severely restrict DNAS methods in industrial scenarios: 1) vibration signals are multi-scale and non-stationary; 2) huge memory cost by supernet-based search is unsuitable to practical diagnosis; 3) manual architecture derivation causes performance collapse between architecture search and practical diagnosis. Thus, we propose Differentiable Sampling based Efficient Architecture Search (DS-EAS), which generates architecture by differentiable sampling. First, the operator involution is introduced to adaptively extract critical features from noisy signals. Second, Gumbel Max-Softmax is adopted to forward sample and backward propagate the gradient on single sub-architecture at one iteration, alleviating huge memory cost. Third, progressively pruning is proposed to eliminate manual discretization error, leading to the final architecture with zero operators. Based on the searched architecture, a deeper one is built to test its real performance. Traction motor experiment is performed to discuss the performance of DS-EAS on three different sample cases. Compared with other state-of-the-art methods, outperformance of DS-EAS is successfully verified.

Contour subregion error detection methodology using deep learning auto-segmentation.

Inaccurate manual organ delineation is one of the high-risk failure modes in radiation treatment. Numerous automated contour quality assurance (QA) systems have been developed to assess contour acceptability; however, manual inspection of flagged cases is a time-consuming and challenging process, and can lead to users overlooking the exact error location. Our aim is to develop and validate a contour QA system that can effectively detect and visualize subregional contour errors, both qualitatively and quantitatively. A novel contour subregion error detection (CSED) system was developed using subregional surface distance discrepancies between manual and deep learning auto-segmentation (DLAS) contours. A validation study was conducted using a head and neck public dataset containing 339 cases and evaluated according to knowledge-based pass criteria derived from a clinical training dataset of 60 cases. A blind qualitative evaluation was conducted, comparing the results from the CSED system with manual labels. Subsequently, the CSED-flagged cases were re-examined by a radiation oncologist. The CSED system could visualize the diverse types of subregional contour errors qualitatively and quantitatively. In the validation dataset, the CSED system resulted in true positive rates (TPR) of 0.814, 0.800, and 0.771; false positive rates (FPR) of 0.310, 0.267, and 0.298; and accuracies of 0.735, 0.759, and 0.730, for brainstem and left and right parotid contours, respectively. The CSED-assisted manual review caught 13 brainstem, 19 left parotid, and 21 right parotid contour errors missed by conventional human review. The TPR/FPR/accuracy of the CSED-assisted manual review improved to 0.836/0.253/0.784, 0.831/0.171/0.830, and 0.808/0.193/0.807 for each structure, respectively. Further, the time savings achieved through CSED-assisted review improved by 75%, with the time for review taking 24.81±12.84, 26.75±10.41, and 28.71±13.72s for each structure, respectively. The CSED system enables qualitative and quantitative detection, localization, and visualization of manual segmentation subregional errors utilizing DLAS contours as references. The use of this system has been shown to help reduce the risk of high-risk failure modes resulting from inaccurate organ segmentation.

Congruent Weber-B Ankle Fractures do not affect Tibiotalar Contact Mechanics: No Need for the Scalpel?

Category: Trauma; Ankle Introduction/Purpose: Weber-B ankle fractures represent an increasingly common injury world-wide, and the decision for operative fixation often hinges on the stability and congruency of the ankle. While tibiotalar displacement is typically evaluated using the medial clear space on plain radiographs, this method remains prone to inaccuracy because of x-ray beam rotation and manual measurement errors. Furthermore, the influence of these fractures on the mechanical environment of the ankle remains unknown. The recent advent of weightbearing cone-beam CT (WBCT) overcomes these drawbacks by imaging during bipedal stance, allowing a 3D anatomical and mechanical analysis. Therefore, the aim of this study was to analyze the 3D mortise displacement and contact mechanics in weber B ankle fractures by use of WBCT in comparison to their healthy contralateral side. Methods: In this retrospective study, our ankle trauma database was searched for Weber-B ankle fractures imaged by bilateral WBCT imaging between 2015 and 2022. Exclusion criteria consisted of metal or motion artifacts, presence of ankle osteoarthritis, and an age less than 18 years or more than 75 years. Segmentation into 3D models of bone was performed semi-automatically, while personalized cartilage layers were modeled based on a previously validated methodology. Bilateral ankle models were imported in custom-made Matlab® script for an automated anatomical and mechanical analysis. 3D mortise congruency was evaluated by use of following parameters: fibular length, talocrural angle, distance mapping of the medial gutter and tibiofibular clear space distance mapping. Contact mechanics were evaluated by the mean and maximum contact stress of the tibia and talus, as well as the contact area (Figure 1). A student’s T-test was performed to investigate the difference between the fractured and healthy side. Results: Thirty-two patients, with mean age 38.50 years (SD = 15.81 ) and weight 79.12 kg (SD = 16,24), were confirmed eligible for analysis. Statistical analysis revealed that there were no significant differences for all anatomical parameters (P &gt; 0.05). The mean contact stress of the fractured and healthy side was 2.10 (SD = 0.42) MPa and 2.10 (SD = 0,41) MPa, respectively, whereas the maximum stress was 7.67 MPa (SD = 1.55) and 7,47 MPa (SD = 1,67), respectively. No statistical significant differences were found between all mechanical parameters (P &gt; 0.05). Conclusion: This study demonstrates that contact mechanics are not affected in Weber-B fractures with a congruent mortise. Therefore, non-operative treatment could be considered, as there will theoretically be no increased risk for the patients to develop posttraumatic osteoarthritis in the long term. In clinical practice, WBCT could prove useful to determine the 3D anatomical and mechanical environment of ankle fractures and guide patients towards (non-) operative treatment. However, further studies should focus on the minimal clinical important difference/rehabilitation factors associated with mortise malalignment based on concomitant contact stress increase, which would warrant and identify which cases need surgical reduction.