Baseline Accuracy Research Articles

Spectral graph convolutional neural network for Alzheimer's disease diagnosis and multi-disease categorization from functional brain changes in magnetic resonance images

Alzheimer's disease (AD) is a progressive neurological disorder characterized by the gradual deterioration of cognitive functions, leading to dementia and significantly impacting the quality of life for millions of people worldwide. Early and accurate diagnosis is crucial for the effective management and treatment of this debilitating condition. This study introduces a novel framework based on Spectral Graph Convolutional Neural Networks (SGCNN) for diagnosing AD and categorizing multiple diseases through the analysis of functional changes in brain structures captured via magnetic resonance imaging (MRI). To assess the effectiveness of our approach, we systematically analyze structural modifications to the SGCNN model through comprehensive ablation studies. The performance of various Convolutional Neural Networks (CNNs) is also evaluated, including SGCNN variants, Base CNN, Lean CNN, and Deep CNN. We begin with the original SGCNN model, which serves as our baseline and achieves a commendable classification accuracy of 93%. In our investigation, we perform two distinct ablation studies on the SGCNN model to examine how specific structural changes impact its performance. The results reveal that Ablation Model 1 significantly enhances accuracy, achieving an impressive 95%, while Ablation Model 2 maintains the baseline accuracy of 93%. Additionally, the Base CNN model demonstrates strong performance with a classification accuracy of 93%, whereas both the Lean CNN and Deep CNN models achieve 94% accuracy, indicating their competitive capabilities. To validate the models' effectiveness, we utilize multiple evaluation metrics, including accuracy, precision, recall, and F1-score, ensuring a thorough assessment of their performance. Our findings underscore that Ablation Model 1 (SGCNN Model 1) delivers the highest predictive accuracy among the tested models, highlighting its potential as a robust approach for Alzheimer's image classification. Ultimately, this research aims to facilitate early diagnosis and treatment of AD, contributing to improved patient outcomes and advancing the field of neurodegenerative disease diagnosis.

A deep learning framework for automated anomaly detection and localization in fused filament fabrication

Under-extrusion poses a common challenge in filament 3D printing, necessitating precise adjustments to printing parameters for optimal resolution. Swift identification of this flaw is crucial for implementing timely corrective measures. In response, we present a novel machine learning approach designed to detect anomalies in 3D printing, specifically in the fused filament fabrication process (FFF). Our framework utilizes “You Only Look Once” (YOLO), a real-time object detection system, and “Visual Geometry Group-16” (VGG-16), a Convolutional Neural Network (CNN) model for image recognition, to accurately identify and localize under-extrusion events. Initially, models such as VGG-16, VGG-19, and ResNet-50 were trained without the inclusion of YOLO to establish baseline accuracies. Subsequently, an automated image pre-processing phase employs YOLO to discern the nozzle head, facilitating subsequent cropping around the region of interest for retraining the models. This inclusion of the nozzle head detection approach significantly improved the accuracy of all models, with the combined application of YOLO and VGG-16 demonstrating the most substantial enhancement, boosting detection accuracy to 97%. Moreover, Gradient-weighted Class Activation Mapping (Grad-CAM) technique, another CNN-based method, is employed to effectively highlight predicted areas in under-extrusion scenarios. While our method significantly advances the automatic detection of printing anomalies, it primarily serves as a diagnostic tool, signaling the need for intervention. Our rigorous testing on images of varying complexities confirms the model’s robustness, evaluated using comprehensive metrics such as precision, recall, and F1 score.

Hardening behavioral classifiers against polymorphic malware: An ensemble approach based on minority report

In recent years, malware attacks have become more and more sophisticated, reflecting a radical change in malware behavior. Attackers aim to create malware that, at each execution, generates a different number of independent and cooperating threads. Randomization of malware's division of labor among threads poses significant challenges to traditional detection approaches. In this paper, we demonstrate that attacks based on random division of labor among multiple threads can dramatically degrade the detection performance of five benchmark ML models, in some cases dropping their accuracy to 50% with only a few threads. Then, we propose and evaluate a novel detection technique based on polymorphic-aware training and ensemble learning with ad-hoc voting scheme (favoring minority report). Results of experimentation carried out on real malware system call logs and assigned to threads via a Bayesian splitting accounting for inter-call dependency indicate that our ensemble has high detection capabilities (99.7% best case), and improves the baseline accuracy of a single model in detecting single-thread malware.

Optimizing the Delivery of Routine Malaria Data Quality Assessments: A Two-Level Logistic Regression Model to Inform an Institutionalized Approach in Mozambique.

Mozambique has implemented routine data quality assessments (DQAs) to improve accuracy of health facility (HF) malaria reporting since 2019. However, despite this being a resource-intensive exercise, the impact of operational factors on DQAs has not yet been systematically investigated. This analysis aimed to provide insights into optimizing the operational delivery of routine DQAs. A two-level logistic regression model based on 1,354 DQAs conducted across 195 HFs (16 districts, November 2019-December 2022) was used to estimate the impact of relevant operational factors, namely number of DQAs received, baseline reporting accuracy, HF setting, workload, malaria transmission intensity, and the shift to digital reporting, on accurate reporting by HFs. A report was considered accurate if the deviation between number of confirmed malaria cases in reports and register books was less than 10%. A statistically significant interaction was observed between baseline reporting accuracy and number of DQAs. For HFs with a baseline accuracy of ≤90%, each additional DQA increased the odds of accurate reporting by 102.8% (95% CI: 71.1-140.2%). For HFs with inaccurate data at baseline, the probability of accurate reporting increased to >80% after five DQAs, whereas HFs with accurate baseline data did not improve beyond the baseline visit. Other operational factors did not significantly affect reporting accuracy. Prioritizing HFs with low baseline accuracy for more frequent DQAs (every 6 months) with at least one visit to all HFs every 3 years might optimize resource allocation in Mozambique. Similar analytic approaches can be applied in other countries to optimize resource allocations for the delivery of routine DQAs.

Adapting magnetoresistive memory devices for accurate and on-chip-training-free in-memory computing.

Memristors have emerged as promising devices for enabling efficient multiply-accumulate (MAC) operations in crossbar arrays, crucial for analog in-memory computing (AiMC). However, variations in memristors and associated circuits can affect the accuracy of analog computing. Typically, this is mitigated by on-chip training, which is challenging for memristors with limited endurance. We present a hardware-software codesign using magnetic tunnel junction (MTJ)-based AiMC off-chip calibration that achieves software accuracy without costly on-chip training. Hardware-wise, MTJ devices exhibit ultralow cycle-to-cycle variations, as experimentally evaluated over 1 million mass-produced devices. Software-wise, leveraging this, we propose an off-chip training method to adjust deep neural network parameters, achieving accurate AiMC inference. We validate this approach with MAC operations, showing improved transfer curve linearity and reduced errors. By emulating large-scale neural network models, our codesigned MTJ-based AiMC closely matches software baseline accuracy and outperforms existing off-chip training methods, highlighting MTJ's potential in AI tasks.

3D whole body preclinical micro-CT database of subcutaneous tumors in mice with annotations from 3 annotators

A pivotal animal model for development of anticancer molecules is mice with subcutaneous tumors, grown by injection of xenografted tumor cells, where micro-Computed Tomography (µCT) of the mice is used to analyze the efficacy of the anticancer molecule. Manual delineation of the tumor region is necessary for the analysis, which is time-consuming and inconsistent, highlighting the need for automatic segmentation (AS) tools. This study introduces a preclinical µCT database, comprising 452 whole-body scans from 223 individual mice with subcutaneous tumors, spanning ten diverse µCT datasets conducted between 2014 and 2020 on a preclinical PET/CT scanner, making it the hitherto largest dataset of its kind. Each tumor is annotated manually by three expert annotators, allowing for robust model development. Inter-annotator agreement was analyzed, and we report an overall annotation agreement of 0.903 ± 0.046 (mean ± std) Fleiss’ Kappa and a mean deviation in volume estimation of 0.015 ± 0.010 cm3 (6.9% ± 4.7), which establishes a human baseline accuracy for delineation of subcutaneous tumors, while showing good inter-annotator agreement.

Accuracy of clear aligners in the orthodontic rotational movement using different attachment configurations.

To evaluate the accuracy of dental rotational movements using clear aligners with different attachment configurations. This retrospective study analysed 212 teeth from 89 patients undergoing Invisalign treatment. Digital models were analysed after the virtual treatment plan (ST1) and after the first treatment phase (ET1) to evaluate the effective clinical rotational movement. The rotational movements of incisors, canines, and bicuspids were measured using data from the Clincheck Movements Table. ST1 and ET1 were compared to determine the actual rotational movement achieved (ST1-ET1). The presence or absence of attachments (rectangular or optimized) on teeth was analysed. The accuracy of rotational movements among attachment types was compared using the Kruskal-Wallis test. Multiple linear regressions were conducted with accuracy as the dependent variable and tooth type, gender, and age as predictors. Optimized attachments had the highest median accuracy (70%), followed by rectangular (65%), and without attachment (63%), with no significant differences (p = .5). There were no significant differences across age groups, genders, or tooth types. Baseline accuracy was 68.62% (95% CI: 56.03-81.20, p < .001). Age was a significant predictor (estimate = -0.30, 95% CI: -0.58 - -0.03, p = .032), indicating decreased accuracy with increasing age. The model's R2 was 0.046, with an adjusted R2 of 0.003, indicating minimal variance explained. The addition of attachment configurations to clear aligners improves rotational accuracy, but not significantly. Further advancements in these configurations are needed to enhance the performance of the aligners.

Feature Enhanced Spatial–Temporal Trajectory Similarity Computation

AbstractTrajectory similarity computation is a fundamental function in many applications of urban data analysis, such as trajectory clustering, trajectory compression, and route planning. In this paper, we study trajectory similarity computation on the road network. However, existing methods have been designed primarily for road network trajectories with spatial information, while ignoring the important temporal information in the real world. To solve this problem, we propose a Feature Enhanced Spatial–Temporal trajectory similarity computation framework FEST, which is a graph neural network (GNN) and sequence model pipeline. We first use the GNN model to capture global information on the road network. In particular, we enhance the process with multi-graph to learn multiple signals from the road network on different aspects. In addition to the original road network topology signal, we also take into account the content signal to learn spatial–temporal features from trajectory traffic, as well as the adaptive similarity signal of the road network to learn hidden features. From these three signals, we construct a multi-graph and use GCN to learn road intersection embedding jointly. Next, we propose a feature-enhanced Transformer with spatial–temporal information to learn correlation within the trajectory, and we further use mean-pooling to get the final trajectory embedding. We compare FEST with six trajectory similarity computation methods on two real-world datasets. The results show that FEST consistently outperforms all baselines and can improve the accuracy of the best-performing baseline.

Investigation of the accuracy of BeiDou, QZSS and QZSS/BeiDou satellites configuration for short, medium and long baselines in the Asia-Pacific regions

Abstract The field of satellite navigation has seen significant advancements due to the fast development of multi-constellation Global Navigation Satellite Systems (GNSS). Around 150 satellites will be in service when all six systems – GPS, GLONASS, Galileo, BeiDou, QZSS, and NAVIC – are launched by 2030, offering both enormous potential and advantages for research and engineering applications. This study used an experiment on the accuracy, particularly for short, medium, long baselines (Wide Lane ambiguity solution) of the BeiDou, QZSS and QZSS/BeiDou combinations. It showed that with the integration of BeiDou/QZSS static measurements in the study region millimetre-centimetre accuracy for short, medium, and long baselines can be attained. Based on the results of this study, it can be concluded that the 1st (QZSS/BeiDou), 2nd (BeiDou), and 3rd (QZSS) strategies feature different horizontal accuracies for all categories. The obtained results with different satellite configurations for the Fixed-Wide-Lane integer ambiguity solution are compared with each other. Accuracy at the short baseline (BeiDou, QZSS, and BeiDou/QZSS satellites) was obtained in the range of 0.5–0.7 cm. For the medium baseline, it was computed around 1.8–82 cm. For the long baseline, the accuracy was 5.6–13.3 cm.

Interferometric Calibration Model for the LuTan-1 Mission: Enhancing Digital Elevation Model Accuracy

The LuTan-1 (LT-1) mission, China’s first civilian bistatic spaceborne Synthetic Aperture Radar (SAR) mission, comprises two L-band SAR satellites. These satellites operate in bistatic InSAR strip map mode, maintaining a formation flight with an adjustable baseline to generate global digital elevation models (DEMs) with high accuracy and spatial resolution. This research introduces a dedicated interferometric calibration model for LT-1, tackling the unique challenges of the bistatic system, such as interferometric parameter coupling and the π-ambiguity problem caused by synchronization phase errors. This study validates the model using SAR images from LT-1 and Xinjiang corner reflector data, achieving interferometric phase accuracy better than 0.1 rad and baseline accuracy better than 2 mm, thereby producing high-precision DEMs with a height accuracy meeting the 5 m requirement.

Accuracy assessment of GEDI terrain elevation, canopy height, and aboveground biomass density estimates in Japanese artificial forests

Global forests face severe challenges owing to climate change, making dynamic and accurate monitoring of forest conditions critically important. Forests in Japan, covering approximately 70% of the country's land area, play a vital role yet often overlooked in global forestry. Japanese forests are unique, with approximately 50% comprising artificial forests, predominantly coniferous forests. Despite the Japanese government's extensive use of airborne Light Detecting and Ranging (LiDAR) to assess forest conditions, these data need more availability and frequency. The Global Ecosystem Dynamics Investigation (GEDI), the first Spaceborne LiDAR data explicitly designed for vegetation monitoring, is expected to provide significant value for high-frequency and high-accuracy forest monitoring. To assess the accuracy of GEDI data in Japanese artificial coniferous forests, the reference data were gathered from 53,967,770 artificial coniferous trees via airborne LiDAR data in Aichi Prefecture, Japan. This data was then compared to the corresponding GEDI-derived terrain elevations, canopy heights (GEDI RH98), and aboveground biomass density (AGBD) estimates to assess the accuracy of GEDI data. This research also explored how different factors influence the accuracy of GEDI terrain elevation estimates, including the type of beam, time of acquisition (day or night), beam sensitivity, and terrain slope. Additionally, the effects of various forest structural parameters, such as the height-to-diameter ratio, crown length ratio, and the number of trees on the accuracy of the GEDI canopy height and AGBD, were investigated. The results showed that GEDI terrain elevation demonstrated high accuracy across various slope conditions, with rRMSE ranging from 2.28% to 3.25% and RMSE ranging from 11.68 m to 16.54 m. After geolocation adjustment, the comparison of canopy height estimates derived from GEDI to airborne LiDAR-derived canopy height also showed high accuracy, exhibiting a rRMSE of 22.04%. In contrast, the GEDI AGBD product showed moderate accuracy, with a rRMSE of 52.79%. The findings also indicated that the accuracy of GEDI RH98 was influenced by terrain slope and crown length ratio, whereas the accuracy of GEDI AGBD was mainly impacted by the number of trees and crown length ratio. This study provided the first baseline accuracy assessment of GEDI terrain elevation, RH98, and AGBD estimates in Japanese artificial forests. Furthermore, this study provided valuable insights into the accuracy of GEDI metrics by examining potential factors.

A differential-based method for correcting ultra-short baseline positioning results

The primary factors influencing the positional accuracy of an ultra-short baseline (USBL) system are errors in sound speed, delay, installation deviation, and the bending of sound rays. Premised on the USBL positioning model, an error contribution model including ranging error, delay error, sound speed error and array element installation deviation was constructed. In parallel, an indirect difference correction method with additional coefficients was proposed based on the geometric correlation of the error function model. Specifically, a reference beacon was deployed on the seabed, and differential correction was achieved by directly computing target coordinates, integrating them with the positional data from the reference beacon. The proposed method demonstrated efficacy in mitigating errors associated with gradual variations or pronounced geometric correlations, including discrepancies in sound speed, bending of sound rays, and deviations in the installation of array elements. The correction method was applied to the USBL positioning, and the simulation results show that the proposed method notably enhanced both accuracy and consistency in positioning, particularly in shallow water environments characterised by significant sound speed errors. The sea trial data from the R/V Dayang Yihao in the Western Pacific demonstrate that the proposed method can effectively mitigate the errors induced by sound ray bending in deep-sea environments. The mean absolute error (MAE) and standard deviation (STD) were reduced by 51.3% and 55.3%, respectively, compared with the results of the USBL direct solution.

Few-shot classification of ultrasound breast cancer images using meta-learning algorithms

Medical datasets often have a skewed class distribution and a lack of high-quality annotated images. However, deep learning methods require a large amount of labeled data for classification. In this study, we present a few-shot learning approach for the classification of ultrasound breast cancer images using meta-learning methods. We used prototypical networks and model agnostic meta-learning (MAML) algorithms as meta-learning methods. The breast ultrasound images (BUSI) dataset, which has three classes and is difficult to use in meta-learning, was used for meta-testing in a cross-domain approach along with other datasets for meta-training. Our proposed approach yielded an accuracy range of 0.882–0.889, achieved by implementing the ResNet50 backbone with ProtoNet in a 10-shot setting. These results represent a significant improvement ranging from 6.27 to 7.10% over the baseline accuracy of 0.831. The results showed that ProtoNet outperformed the MAML method for all k-shot settings. In addition, the use of ResNet models as the backbone network for feature extraction was found to be more successful than the use of a four-layer convolutional model. Our proposed method is the first attempt to apply meta-learning for few-shot classification in the BUSI dataset while providing higher accuracy compared to deep learning methods for medical images with small-scale datasets and few classes. The methodology used in this study can be adapted to other datasets with similar problems.

Sensorimotor Network Segregation Predicts Long-Term Learning of Writing Skills in Parkinson's Disease.

The prediction of motor learning in Parkinson's disease (PD) is vastly understudied. Here, we investigated which clinical and neural factors predict better long-term gains after an intensive 6-week motor learning program to ameliorate micrographia. We computed a composite score of learning through principal component analysis, reflecting better writing accuracy on a tablet in single and dual task conditions. Three endpoints were studied-acquisition (pre- to post-training), retention (post-training to 6-week follow-up), and overall learning (acquisition plus retention). Baseline writing, clinical characteristics, as well as resting-state network segregation were used as predictors. We included 28 patients with PD (13 freezers and 15 non-freezers), with an average disease duration of 7 (±3.9) years. We found that worse baseline writing accuracy predicted larger gains for acquisition and overall learning. After correcting for baseline writing accuracy, we found female sex to predict better acquisition, and shorter disease duration to help retention. Additionally, absence of FOG, less severe motor symptoms, female sex, better unimanual dexterity, and better sensorimotor network segregation impacted overall learning positively. Importantly, three factors were retained in a multivariable model predicting overall learning, namely baseline accuracy, female sex, and sensorimotor network segregation. Besides the room to improve and female sex, sensorimotor network segregation seems to be a valuable measure to predict long-term motor learning potential in PD.

Real-Time Motorbike Detection: AI on the Edge Perspective

Motorbikes are an integral part of transportation in emerging countries, but unfortunately, motorbike users are also one the most vulnerable road users (VRUs) and are engaged in a large number of yearly accidents. So, motorbike detection is very important for proper traffic surveillance, road safety, and security. Most of the work related to bike detection has been carried out to improve accuracy. If this task is not performed in real-time then it loses practical significance, but little to none has been reported for its real-time implementation. In this work, we have looked at multiple real-time deployable cost-efficient solutions for motorbike detection using various state-of-the-art embedded edge devices. This paper discusses an investigation of a proposed methodology on five different embedded devices that include Jetson Nano, Jetson TX2, Jetson Xavier, Intel Compute Stick, and Coral Dev Board. Running the highly compute-intensive object detection model on edge devices (in real-time) is made possible by optimization. As a result, we have achieved inference rates on different devices that are twice as high as GPUs, with only a marginal drop in accuracy. Secondly, the baseline accuracy of motorbike detection has been improved by developing a custom network based on YoloV5 by introducing sparsity and depth reduction. Dataset augmentation has been applied at both image and object levels to enhance robustness of detection. We have achieved 99% accuracy as compared to the previously reported 97% accuracy, with better FPS. Additionally, we have provided a performance comparison of motorbike detection on the different embedded edge devices, for practical implementation.

Phishing URL Detection using Machine Learning

Abstract: Phishing attacks pose a significant threat in cyberspace, Cybercrimes such as Phishing means accessing personal data and violating security through the Internet and its main aim is to steal the information from the users using different techniques in that the primary one is Phishing which demands effective detection mechanisms. This study evaluates the performance of Gradient Boosting Classifier, Random Forest, and Decision Tree machine learning models in conjunction with feature selection techniques namely SelectKBest and Chi-Square. Initially, a comprehensive feature set of 30 attributes achieved a baseline accuracy of 97.4%. Through SelectKBest and Chi-Square feature selection methods, 13 key features were identified, leading to a slightly decreased accuracy of 95.6% upon model retraining. This research highlights the importance of feature selection in enhancing phishing detection accuracy while maintaining model interpretability. Technologies such as NumPy, Pandas, Matplotlib, Scikit-learn, and Flask drive. The project, emphasising the exploration of ML models, EDA (Exploratory Data Analysis) on phishing data, and understanding feature importance. The Machine Learning Models like Gradient Boosting Classifier, Random Forest and Decision tree are used to detect whether the given URL is Malicious URL or Legitimate URL using Comprehensive Feature set and Key Feature set. This Models and Feature sets are compared based on the Performance Metrics namely Accuracy, Precision, Recall and F1- score. This study contributes to advancing cyber defence mechanism through the fusion of sophisticated Machine Learning algorithms and meticulous Feature Selection methodologies.

Shrinking Your TimeStep: Towards Low-Latency Neuromorphic Object Recognition with Spiking Neural Networks

Neuromorphic object recognition with spiking neural networks (SNNs) is the cornerstone of low-power neuromorphic computing. However, existing SNNs suffer from significant latency, utilizing 10 to 40 timesteps or more, to recognize neuromorphic objects. At low latencies, the performance of existing SNNs is drastically degraded. In this work, we propose the Shrinking SNN (SSNN) to achieve low-latency neuromorphic object recognition without reducing performance. Concretely, we alleviate the temporal redundancy in SNNs by dividing SNNs into multiple stages with progressively shrinking timesteps, which significantly reduces the inference latency. During timestep shrinkage, the temporal transformer smoothly transforms the temporal scale and preserves the information maximally. Moreover, we add multiple early classifiers to the SNN during training to mitigate the mismatch between the surrogate gradient and the true gradient, as well as the gradient vanishing/exploding, thus eliminating the performance degradation at low latency. Extensive experiments on neuromorphic datasets, CIFAR10-DVS, N-Caltech101, and DVS-Gesture have revealed that SSNN is able to improve the baseline accuracy by 6.55% ~ 21.41%. With only 5 average timesteps and without any data augmentation, SSNN is able to achieve an accuracy of 73.63% on CIFAR10-DVS. This work presents a heterogeneous temporal scale SNN and provides valuable insights into the development of high-performance, low-latency SNNs.

Analyzing and Improving Existing Neural Network-Based Approaches to Identify AI Generated Images

Images produced by AI software, particularly human faces, have the potential to spread misinformation, thus detection tools are essential to tell facts from fiction. However, current neural network-based tools misclassify images with extreme photography conditions, facial obstructions, or post-processing. This project aimed to address the limitations of existing detection models, especially regarding poorly-photographed or edited images. A MobileNet-v2-based neural network to identify synthetic images was analyzed for possible shortcomings. After obtaining its baseline accuracy, it was stress tested on validation data with brightness, contrast, hue, and saturation edited by set increments. The model performed poorly on processed images and faces with glasses or shadows. Possible limitations were insufficient data augmentation, the Global Average Pooling Layer, too few epochs trained, and inadequate regularization strength or dropout rate. The model was then modified to address these shortcomings by altering the brightness, contrast, hue, and saturation of training data, adding Gaussian noise to training images, implementing a Global Max Pooling Layer, augmenting the number of epochs trained, and changing the regularization strength and dropout rate. The modified model was evaluated on augmented and non-augmented images. Overall, the model improved when Gaussian noise was added with standard deviation 0.1, it was trained for 15 epochs, and there were no dropout layers. This was because the Gaussian noise layer approximated degraded image quality, and the model was too simple to learn image features when trained for fewer epochs or when neurons were disabled. This research could be expanded by testing multiple modifications at once.

Deep Learning Glioma Grading with the Tumor Microenvironment Analysis Protocol for Comprehensive Learning, Discovering, and Quantifying Microenvironmental Features.

Gliomas are primary brain tumors that arise from neural stem cells, or glial precursors. Diagnosis of glioma is based on histological evaluation of pathological cell features and molecular markers. Gliomas are infiltrated by myeloid cells that accumulate preferentially in malignant tumors, and their abundance inversely correlates with survival, which is of interest for cancer immunotherapies. To avoid time-consuming and laborious manual examination of images, a deep learning approach for automatic multiclass classification of tumor grades was proposed. As an alternative way of investigating characteristics of brain tumor grades, we implemented a protocol for learning, discovering, and quantifying tumor microenvironment elements on our glioma dataset. Using only single-stained biopsies we derived characteristic differentiating tumor microenvironment phenotypic neighborhoods. The study was complicated by the small size of the available human leukocyte antigen stained on glioma tissue microarray dataset - 206 images of 5 classes - as well as imbalanced data distribution. This challenge was addressed by image augmentation for underrepresented classes. In practice, we considered two scenarios, a whole slide supervised learning classification, and an unsupervised cell-to-cell analysis looking for patterns of the microenvironment. In the supervised learning investigation, we evaluated 6 distinct model architectures. Experiments revealed that a DenseNet121 architecture surpasses the baseline's accuracy by a significant margin of 9% for the test set, achieving a score of 69%, increasing accuracy in discerning challenging WHO grade 2 and 3 cases. All experiments have been carried out in a cross-validation manner. The tumor microenvironment analysis suggested an important role for myeloid cells and their accumulation in the context of characterizing glioma grades. Those promising approaches can be used as an additional diagnostic tool to improve assessment during intraoperative examination or subtyping tissues for treatment selection, potentially easing the workflow of pathologists and oncologists.

The influence of the da Vinci surgical robot on electromagnetic tracking in a clinical environment

Robot-assisted surgery is increasingly used in surgery for cancer. Reduced overview and loss of anatomical orientation are challenges that might be solved with image-guided surgical navigation using electromagnetic tracking (EMT). However, the robot’s presence may distort the electromagnetic field, affecting EMT accuracy. The aim of this study was to evaluate the robot’s influence on EMT accuracy. For this purpose, two different electromagnetic field generators were used inside a clinical surgical environment: a table top field generator (TTFG) and a planar field generator (PFG). The position and orientation of sensors within the electromagnetic field were measured using an accurate in-house developed 3D board. Baseline accuracy was measured without the robot, followed by stepwise introduction of potential distortion sources (robot and robotic instruments). The absolute accuracy was determined within the entire 3D board and in the clinical working volume. For the baseline setup, median errors in the entire tracking volume within the 3D board were 0.9 mm and 0.3° (TTFG), and 1.1 mm and 0.4° (PFG). Adding the robot and instruments did not affect the TTFG’s position accuracy (p = 0.60), while the PFG’s accuracies decreased to 1.5 mm and 0.7° (p < 0.001). For both field generators, when adding robot and instruments, accuracies inside the clinical working volume were higher compared to the entire tracking 3D board volume, 0.7 mm and 0.3° (TTFG), and 1.1 mm and 0.7° (PFG). Introduction of a surgical robot and robotic instruments shows limited distortion of the EMT field, allowing sufficient accuracy for surgical navigation in robotic procedures.