Pivotal Procedure Research Articles

Multi-stream and multi-scale fusion rib fracture segmentation network based on UXNet.

Accurate segmentation of rib fractures represents a pivotal procedure within surgical interventions. This meticulous process not only mitigates the likelihood of postoperative complications but also facilitates expedited patient recuperation. However, rib fractures in computed tomography (CT) images exhibit an uneven morphology and are not fixed in position, posing difficulties in segmenting fractures. This study aims to enhance the accuracy of elongated rib fracture segmentation, ultimately improving the efficiency of clinical diagnosis. In this study, we propose multi-stream and multi-scale fusion network based on efficient attention UXNet (M2SUXNet). It aims to enhance the segmentation accuracy of elongated rib fractures through multi-scale fusion attention enhancement. Firstly, we propose the multi-stream and multi-scale fusion (M2SF) module in the feature extraction stage. The module is designed with two parallel paths. Each path analyzes the image content using a different feature level. Then, the module effectively distinguishes the more critical feature information in the channel according to the feature weight ratio. The M2SF module integrates information from different scales to obtain comprehensive information on global and local features, achieving a more diverse feature representation. Secondly, the efficient attention (EA) module combines different channel information of input features to integrate channel and spatial features of different channels. The module better combines the context information, establishes the dependency between the space and the channel, enhances the focusing ability of the network on the fractures of different shapes, and improves the segmentation accuracy. Thirdly, the joint loss function of BCE with Logits Loss and Dice Loss is used to solve the sample imbalance problem. We verified the effectiveness of the proposed model on the public RibFrac dataset. The experimental results demonstrated that the model achieved a Dice coefficient of 75.34%, a joint intersection over union (IoU) of 60.44%, and a precision of 93.79%. The proposed model for rib fracture segmentation has higher accuracy and feasibility than other existing models. Besides, the M2SUXNet can effectively improve the segmentation performance of elongated rib fractures.

Endoscopic Ultrasound Assessment of Solid Pseudopapillary Neoplasm of the Pancreas: Case Series in Latin American Population

Introduction: Solid pseudopapillary tumors of the pancreas (SPN) are uncommon lesions. Endoscopic ultrasound is considered the standard examination because of its capability to acquire a suitable core tissue sample. This study details the experience of eleven cases within a Latin American population diagnosed with SPNs through the endoscopic ultrasound-guided fine needle biopsy (EUS-FNB) approach. Methods: Retrospective reviews of records from a 6-year period, from January 2018 to December 2023, were performed at Clínica Reina Sofía in Bogotá, Colombia. We included eleven patients with surgically proven solid pseudopapillary tumors who had undergone preoperative EUS−FNB. The clinical history, EUS findings, biopsies, and immunohistochemical profile were reviewed. Results: In this study of eleven SPN patients (median age: 31.9 years, 90.1% female, all Hispanic), abdominal pain (63.6%) was the predominant symptom. The indication for EUS was predominantly a pancreas-dependent mass (90.1%). Tumors were located more frequently on the pancreatic neck, with regular morphology, well-defined borders, and predominantly hypoechoic and heterogeneous appearances on EUS. The average tumor diameter was 4.3 cm [range 1.2- 10 cm]. Most tumors were solid (63.6%), and elastography revealed a mixed pattern. EUS-Doppler indicated hypovascularity in all cases. Vascular compression occurred in three patients. No lymph nodes were observed. There were no complications related to the procedure. The histopathological analysis using EUS-FNB yielded consistent results with post-surgical biopsies. Conclusion: In the context of diagnostic evaluation for SPNs, EUS-FNB emerges as a pivotal procedure. In this descriptive study, EUS-FNB showed a remarkable preoperative diagnostic yield of 100% compared to post-surgical histopathology for solid pseudopapillary tumors.

Plant sperm cell sequencing for genome phasing and determination of meiotic crossover points.

Haplotype phasing represents a pivotal procedure in genome analysis, entailing the identification of specific genetic variant combinations on each chromosome. Achieving chromosome-level genome phasing constitutes a considerable challenge, particularly in organisms with large and complex genomes. To address this challenge, we have developed a robust, gamete cell-based phasing pipeline, including wet-laboratory processes for plant sperm cell isolation, short-read sequencing and a bioinformatics workflow to generate chromosome-level phasing. The bioinformatics workflow is applicable for both plant and other sperm cells, for example, those of mammals. Our pipeline ensures high-quality single-nucleotide polymorphism (SNP) calling for each sperm cell and the subsequent construction of a high-density genetic map. The genetic map facilitates accurate chromosome-level genome phasing, enables crossover event detection and could be used to correct potential assembly errors. Our bioinformatics pipeline runs on a Linux system and most of its steps can be executed in parallel, expediting the analysis process. The entire workflow can be performed over the course of 1 d. We provide a practical example from our previous research using this protocol and provide the whole bioinformatics pipeline as a Docker image to ensure its easy adaptability to other studies.

A Comprehensive Review of the Endometrial Receptivity Array in Embryo Transfer: Advancements, Applications, and Clinical Outcomes.

Embryo transfer is a pivotal procedure in assisted reproductive technologies (ART). Yet, the success of this process hinges on multiple factors, with endometrial receptivity playing a critical role in determining the likelihood of successful implantation. The endometrial receptivity array (ERA) is an advanced diagnostic tool designed to personalize embryo transfer timing by assessing the endometrium's receptivity. This review comprehensively examines the ERA, exploring its biological foundation, technological development, and clinical applications. The ERA's ability to analyze the expression of genes associated with endometrial receptivity offers a tailored approach to identifying the optimal window of implantation (WOI), particularly benefiting patients with recurrent implantation failure (RIF) or repeated unsuccessful in vitro fertilization (IVF) cycles. Clinical outcomes from ERA-guided embryo transfers indicate improvements in implantation rates and overall pregnancy success, although challenges such as result variability and cost-effectiveness persist. This review also discusses the latest advancements in ERA technology, including integrating genomic and transcriptomic analyses, non-invasive techniques, and using artificial intelligence (AI). Controversies regarding the widespread application of ERA and its necessity in all IVF cases are critically examined. By summarizing the current state of ERA in embryo transfer, this review aims to inform clinicians, researchers, and patients about its potential to enhance ART outcomes and to highlight areas for future research and innovation.

Efficacy and Safety of Obturator Nerve Block in Transurethral Resection of Bladder Tumors: A Comprehensive Review.

Transurethral resection of bladder tumors (TURBT) is a pivotal procedure in the management of bladder cancer, essential for both diagnosis and treatment. Effective anesthesia is crucial in TURBT to ensure a stable and pain-free operative field, facilitate precise tumor resection, and minimize complications such as the obturator reflex, which can lead to involuntary leg movement and bladder injury. The obturator nerve block (ONB) is a regional anesthesia technique designed to prevent the obturator reflex by blocking the obturator nerve, which innervates the adductor muscles of the thigh. This comprehensive review evaluates the efficacy and safety of ONB in TURBT. It begins by discussing the anatomical and physiological aspects of the obturator nerve, followed by a detailed examination of various ONB techniques, including ultrasound-guided and landmark-based methods. The review assesses the impact of ONB on pain management, reduction of adductor muscle spasms, and overall improvement in surgical conditions and patient satisfaction. Additionally, it explores the incidence and types of complications associated with ONB, such as hematoma, nerve injury, and local anesthetic systemic toxicity (LAST). It compares ONB with other anesthesia techniques used in TURBT, such as general, spinal, and epidural anesthesia. A critical analysis of key clinical studies and meta-analyses is presented to provide a comprehensive understanding of the current evidence on ONB efficacy and safety. Future directions and innovations in ONB techniques, including advances in imaging and nerve localization, are also discussed. Practical recommendations for implementing ONB in clinical practice, including guidelines for clinician training and patient selection criteria, are provided. This review aims to inform clinicians about the benefits and risks of ONB in TURBT, guide clinical practice, and identify areas for future research to optimize anesthesia management in bladder cancer surgery.

Non-parametric fragility curves for probabilistic risk assessment of rainfall-triggered landslides

Fragility analysis of slopes under rainfall condition express the conditional probabilities of exceeding prescribed slope limit states under a range of rainfall intensities, which is the basic and pivotal procedure in rainfall-triggered landslide risk assessment. Classical fragility analysis methods of slopes always develop fragility curves with the assumption of lognormal distribution. The validation of assuming the shape of fragility curves as lognormal distribution remains a open question in practice. In addition, the existing fragility analysis of slopes under rainfall condition have not dealt with the effect of spatially-variable geo-material parameters. In the present study, a novel fragility analysis methodology is proposed for the construction of non-parametric fragility curves of slopes under rainfall condition, in which the spatially-variable soil properties are taken into account. The non-parametric fragility curves are developed by the calculated slope failure probabilities under various rainfall intensities on the basis of probability density evolution theory, without restricting fragility functions to some assumed distribution models. With the proposed method and a set of probabilistic slope rainfall seepage and stability analyses, non-parametric fragility curves, fragility surface using two rainfall intensity measures and time-dependent fragility curves of a slope subjected to various rainfall scenarios are obtained for rainfall-triggered landslides risk assessment.

Quantum-Inspired Algorithms for Market Clearing in Smart Grids: A Comprehensive Review

The integration of smart grids into modern power systems marks a fundamental shift, leveraging cutting-edge technology to enhance system efficiency and optimize energy distribution. At the heart of this transformation lies the concept of market clearing, a pivotal procedure facilitating effective energy trading and resource allocation within smart grids. This research explores the significance of market clearing in smart grids and introduces the transformative realm of quantum-inspired algorithms, poised to revolutionize energy market control systems.

Screw Production Optimization Using Artificial Neural Network (ANN) Technology

The application of artificial neural network model and techniques for the optimization of screw production has been investigated. Detection of defects in a company’s production processes is an essential and vital aspect of maintaining quality and smooth production control. The production process can give rise to numerous shortfalls, leading to maximum production expenses. As a result of this, scrutinizing screws, which constitute an important component of various mechanical parts, emerges as a pivotal procedure. This study considered countersunk head 5.2mm screws with stripped screws defects, surface dirt, and surface damage. Industrial camera with efficient specifications for image acquisition of the screws and Deep Convolutional Neural Network (DCNN) which involves software tools (Google’s machine learning architecture Tensorflow, and Keras) technique were utilized for the detection of micro flaws on metal screw surfaces. From the results obtained, the proposed convolutional neural network technique results were compared with traditional machine vision technique such as using LeNet-5 and a detection accuracy of 97% and average detection time per picture is 1.3 s, making the technique faster in processing time and higher in accuracy compare to other validated techniques.

An improved mayfly algorithm based on Kapur entropy for multilevel thresholding color image segmentation

Multilevel thresholding segmentation of color images plays an important role in many fields. The pivotal procedure of this technique is determining the specific threshold of the images. In this paper, an improved mayfly algorithm (IMA)-based color image segmentation method is proposed. Tent mapping initializes the female mayfly population to increase population diversity. Lévy flight is introduced in the wedding dance iterative formulation to make IMA jump from the local optimal solution quickly. Two nonlinear coefficients were designed to speed up the convergence of the algorithm. To better verify the effectiveness, eight benchmark functions are used to test the performance of IMA. The average fitness value, standard deviation, and Wilcoxon rank sum test are used as evaluation metrics. The results show that IMA outperforms the comparison algorithm in terms of search accuracy. Furthermore, Kapur entropy is used as the fitness function of IMA to determine the segmentation threshold. 10 Berkeley images are segmented. The best fitness value, peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and other indexes are used to evaluate the effect of segmented images. The results show that the IMA segmentation method improves the segmentation accuracy of color images and obtains higher quality segmented images.

Aroma-active compounds related to Maillard reaction during roasting in Wuyi Rock tea

Maillard reaction could be a breakthrough in standardizing the roasting process, which is a pivotal procedure in refining Wuyi Rock tea (WRT). This study aimed to find indicators to instruct the roasting process scientifically trying to provide theoretical support for actual production. A total of 86 volatile compounds were detected by solid-phase microextraction (SPME) coupled with gas chromatography–olfactometry–mass spectrometry (GC–O–MS) and GC×GC–O–MS in three rough WRTs leaves (RL) Dahonpao, Shuixian, and Rougui and three roasting degrees, light fire (LF), moderate fire (MF), and high fire (HF). Among them, 45 key aroma-active compounds with r-OAV> 1. Significant differences existed between RL/LF and MF/HF. The Maillard reaction primarily strengthened the roasted attributes but eliminated the individual characteristics through 1-ethyl-1H-pyrrole, 2-ethyl-5-methylpyrazine, 2-ethyl-6-methylpyrazine, 2-ethyl-3-methylpyrazine, furfural, and 5-methyl-2-furancarboxaldehyde. If combined with the analysis of amino acids, sugars, and polyphenols, eight aroma-active compounds relating to the Maillard reaction, furfural, 5-methyl-2-furancarboxaldehyde, 1-ethyl-1H-pyrrole-2-carboxaldehyde, 2-ethyl-5-methylpyrazine, 2-ethyl-6-methylpyrazine, 2,5-dimethyl-3-ethylpyrazine, 1-ethyl-1H-pyrrole, and methylpyrazine, could be considered as indicators to monitor the roasting process. The novelty of this result was that it provides the practical knowledge for further deep investigation from the perspective of the Maillard reaction and makes it possible to industrialize WRT manufacturing, rather than only relying on empiricism.

Two-step artificial intelligence system for endoscopic gastric biopsy improves the diagnostic accuracy of pathologists.

BackgroundEndoscopic biopsy is the pivotal procedure for the diagnosis of gastric cancer. In this study, we applied whole-slide images (WSIs) of endoscopic gastric biopsy specimens to develop an endoscopic gastric biopsy assistant system (EGBAS).MethodsThe EGBAS was trained using 2373 WSIs expertly annotated and internally validated on 245 WSIs. A large-scale, multicenter test dataset of 2003 WSIs was used to externally evaluate EGBAS. Eight pathologists were compared with the EGBAS using a man-machine comparison test dataset. The fully manual performance of the pathologists was also compared with semi-manual performance using EGBAS assistance.ResultsThe average area under the curve of the EGBAS was 0·979 (0·958-0·990). For the diagnosis of all four categories, the overall accuracy of EGBAS was 86·95%, which was significantly higher than pathologists (P< 0·05). The EGBAS achieved a higher κ score (0·880, very good κ) than junior and senior pathologists (0·641 ± 0·088 and 0·729 ± 0·056). With EGBAS assistance, the overall accuracy (four-tier classification) of the pathologists increased from 66·49 ± 7·73% to 73·83 ± 5·73% (P< 0·05). The length of time for pathologists to manually complete the dataset was 461·44 ± 117·96 minutes; this time was reduced to 305·71 ± 82·43 minutes with EGBAS assistance (P = 0·00).ConclusionsThe EGBAS is a promising system for improving the diagnosis ability and reducing the workload of pathologists.

Suboptimal accuracy of ultrasound and ultrasound-based risk stratification systems in detecting medullary thyroid carcinoma should not be overlooked. Findings from a systematic review with meta-analysis.

Ultrasound (US) is the pivotal procedure during the diagnostic work-up of thyroid nodule and several US-based risk stratification systems (RSSs) have been recently developed. Since the performance of RSSs in detecting medullary thyroid carcinoma (MTC) has been rarely investigated, the present systematic review aimed to achieve high evidence about (1) how MTC is classified according to RSSs; (2) if RSSs correctly classify MTC at high risk/suspicion, and (3) if MTC is classified as suspicious at US when RSSs are not used. The review was performed according to MOOSE. The online search was performed by specific algorithm on January 2022. A random-effects model was used for statistical analysis. Twenty-five papers were initially included and their risk of bias was generally low. According to ATA system, 65% of MTCs was assessed at high suspicion and 25% at intermediate suspicion. Considering all RSSs, a 54.8% of MTCs was put in a high-risk/suspicion category. Pooling data from studies without data of RSS the prevalence of ultrasonographically suspicious MTCs was 60%. As conclusion, MTC presentation according to RSSs is partially known and it is classified in a high-risk/suspicion category of RSSs in just over a half of cases. This advises for further studies, ideally supported by international societies, to better define the US presentation of MTC.

Endoscopic retrograde cholangiopancreatography‐related adverse events in Korea: A nationwide assessment

BackgroundAlthough endoscopic retrograde cholangiopancreatography (ERCP) is a pivotal procedure for the diagnosis and treatment of a variety of pancreatobiliary diseases, it has been known that the risk of procedure‐related adverse events (AEs) is significant.ObjectiveWe conducted this nationwide cohort study since there have been few reports on the real‐world data regarding ERCP‐related AEs.MethodsPatients who underwent ERCP were identified between 2012 and 2015 using Health Insurance Review and Assessment database generated by the Korea government. Incidence, annual trends, demographics, characteristics according to the types of procedures, and the risk factors of AEs were assessed.ResultsA total of 114,757 patients with male gender of 54.2% and the mean age of 65.0 ± 15.2 years were included. The most common indication was choledocholithiasis (49.4%) and the second malignant biliary obstruction (22.8%). Biliary drainage (33.9%) was the most commonly performed procedure, followed by endoscopic sphincterotomy (27.4%), and stone removal (22.0%). The overall incidence of ERCP‐related AEs was 4.7% consisting of post‐ERCP pancreatitis (PEP; 4.6%), perforation (0.06%), and hemorrhage (0.02%), which gradually increased from 2012 to 2015. According to the type of procedures, ERCP‐related AEs developed the most commonly after pancreatic stent insertion (11.4%), followed by diagnostic ERCP (5.9%) and endoscopic sphincterotomy (5.7%). Younger age and diagnostic ERCP turned out to be independent risk factors of PEP.ConclusionsERCP‐related AEs developed the most commonly after pancreatic stent insertion, diagnostic ERCP and endoscopic sphincterotomy. Special caution should be used for young patients receiving diagnostic ERCP due to increased risk of PEP.

A Hybrid Preaching Optimization Algorithm Based on Kapur Entropy for Multilevel Thresholding Color Image Segmentation.

Multilevel thresholding segmentation of color images plays an important role in many fields. The pivotal procedure of this technique is determining the specific threshold of the images. In this paper, a hybrid preaching optimization algorithm (HPOA) for color image segmentation is proposed. Firstly, the evolutionary state strategy is adopted to evaluate the evolutionary factors in each iteration. With the introduction of the evolutionary state, the proposed algorithm has more balanced exploration-exploitation compared with the original POA. Secondly, in order to prevent premature convergence, a randomly occurring time-delay is introduced into HPOA in a distributed manner. The expression of the time-delay is inspired by particle swarm optimization and reflects the history of previous personal optimum and global optimum. To better verify the effectiveness of the proposed method, eight well-known benchmark functions are employed to evaluate HPOA. In the interim, seven state-of-the-art algorithms are utilized to compare with HPOA in the terms of accuracy, convergence, and statistical analysis. On this basis, an excellent multilevel thresholding image segmentation method is proposed in this paper. Finally, to further illustrate the potential, experiments are respectively conducted on three different groups of Berkeley images. The quality of a segmented image is evaluated by an array of metrics including feature similarity index (FSIM), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and Kapur entropy values. The experimental results reveal that the proposed method significantly outperforms other algorithms and has remarkable and promising performance for multilevel thresholding color image segmentation.

Adaptive kernel selection network with attention constraint for surgical instrument classification

Computer vision (CV) technologies are assisting the health care industry in many respects, i.e., disease diagnosis. However, as a pivotal procedure before and after surgery, the inventory work of surgical instruments has not been researched with the CV-powered technologies. To reduce the risk and hazard of surgical tools’ loss, we propose a study of systematic surgical instrument classification and introduce a novel attention-based deep neural network called SKA-ResNet which is mainly composed of: (a) A feature extractor with selective kernel attention module to automatically adjust the receptive fields of neurons and enhance the learnt expression and (b) A multi-scale regularizer with KL-divergence as the constraint to exploit the relationships between feature maps. Our method is easily trained end-to-end in only one stage with few additional calculation burdens. Moreover, to facilitate our study, we create a new surgical instrument dataset called SID19 (with 19 kinds of surgical tools consisting of 3800 images) for the first time. Experimental results show the superiority of SKA-ResNet for the classification of surgical tools on SID19 when compared with state-of-the-art models. The classification accuracy of our method reaches up to 97.703%, which is well supportive for the inventory and recognition study of surgical tools. Also, our method can achieve state-of-the-art performance on four challenging fine-grained visual classification datasets.

RETRACTED ARTICLE: Brain Tumor Segmentation Using Deep Learning and Fuzzy K-Means Clustering for Magnetic Resonance Images

The primary objective of this paper is to develop a methodology for brain tumor segmentation. Nowadays, brain tumor recognition and fragmentation is one among the pivotal procedure in surgical and medication planning arrangements. It is difficult to segment the tumor area from MRI images due to inaccessibility of edge and appropriately visible boundaries. In this paper, a combination of Artificial Neural Network and Fuzzy K-means algorithm has been presented to segment the tumor locale. It contains four phases, (1) Noise evacuation (2) Attribute extraction and selection (3) Classification and (4) Segmentation. Initially, the procured image is denoised utilizing wiener filter, and then the significant GLCM attributes are extricated from the images. Then Deep Learning based classification has been performed to classify the abnormal images from the normal images. Finally, it is processed through the Fuzzy K-Means algorithm to segment the tumor region separately. This proposed segmentation approach has been verified on BRATS dataset and produces the accuracy of 94%, sensitivity of 98% specificity of 99%, Jaccard index of 96%. The overall accuracy of this proposed technique has been improved by 8% when compared with K-Nearest Neighbor methodology.

A Dynamic Comparison Sustainability Study of Standard Wastewater Treatment System in the Straw Pulp Papermaking Process and Printing &amp; Dyeing Papermaking Process Based on the Hybrid Neural Network and Emergy Framework

Sustainability study of the standard wastewater treatment system is the pivotal procedure in the water protection field. In order to better study the sustainability of sewage treatment systems, wastewater treatment system of straw pulp papermaking (WTSPP) and wastewater treatment system of printing and dyeing and papermaking (WTPDP) have been selected to assess the sustainable level in China. Based on the hybrid neural network and emergy framework, WTSPP and WTPDP were considered and analyzed in this paper. Therein, three types of indicators were used to evaluate these two systems, including basic structure emergy indicators (BEI), functional emergy indicators (FEI), and eco-efficiency emergy indicators (EEI). Through the basic neural network model and detailed neural network model design, the iteration paths and algorithm operation diagram of WTSPP and WTPDP were designed and realized in this article. Primary contents include: (1) For WTSPP and WTPDP, nonrenewable resources emergy are both the primary contributor and account for roughly 62.5% and 53.7%, respectively. (2) As the important indicator group, the environmental loading ratio (ELR) is 176 in the WTSPP and 323 in the WTPDP, respectively. Emergy sustainability indicators (ESIs) in the WTSPP and WTPDP, are 0.015 and 0.014, respectively. (3) Depending on fluctuation degrees, WTSPP is better than WTPDP. The maximum fluctuation ranges of WTSPP and WTPDP are (3%, −27%) and (28%, 61%), respectively. (4) All neural network analysis results manifest that the emergy sustainability indicators (ESIs) of WTSPP and WTPDP are [0.0151, 0.011] and [0.0179, 0.0055] in view of a long-term predictive view, respectively.

Endoscopic nasopharyngoscopy and ENT specialist safety in the COVID 19 era: the back endoscopy approach to the patient.

PurposeTo propose a way to safely perform endoscopic nasopharyngoscopy in ENT outpatient clinic during the COVID-19 pandemic.MethodsThis manuscript highlights the importance of endoscopy in daily ENT clinical practice, which is a pivotal procedure in the diagnosis of many head and neck pathologies. However, since the beginning of the COVID-19 outbreak, the authors have witnessed a drastic reduction (91%) in the use of endoscopic nasopharyngoscopy at their institutions. In fact, it is considered at risk of contamination for healthcare professionals, as any upper airway manipulation procedure.ResultsIn the “Back approach to the patient” for endoscopic nasopharyngoscopy, the operator positions himself behind the patient and faces the monitor. The endoscopist, not being positioned in front of the patient, should, therefore, be less exposed to airborne transmission of SARS-CoV-2 virus.ConclusionThis simple variation of the physician’s position during endoscopic nasopharyngoscopy could potentially reduce the risk of contagion since the operator is not in the trajectory of droplets and/or aerosols.

Randomised controlled trial of WISENSE, a real-time quality improving system for monitoring blind spots during esophagogastroduodenoscopy

ObjectiveEsophagogastroduodenoscopy (EGD) is the pivotal procedure in the diagnosis of upper gastrointestinal lesions. However, there are significant variations in EGD performance among endoscopists, impairing the discovery rate of gastric cancers...

Lesion location implemented magnetic resonance imaging radiomics for predicting IDH and TERT promoter mutations in grade II/III gliomas

Molecular biological characterization of tumors has become a pivotal procedure for glioma patient care. The aim of this study is to build conventional MRI-based radiomics model to predict genetic alterations within grade II/III gliomas attempting to implement lesion location information in the model to improve diagnostic accuracy. One-hundred and ninety-nine grade II/III gliomas patients were enrolled. Three molecular subtypes were identified: IDH1/2-mutant, IDH1/2-mutant with TERT promoter mutation, and IDH-wild type. A total of 109 radiomics features from 169 MRI datasets and location information from 199 datasets were extracted. Prediction modeling for genetic alteration was trained via LASSO regression for 111 datasets and validated by the remaining 58 datasets. IDH mutation was detected with an accuracy of 0.82 for the training set and 0.83 for the validation set without lesion location information. Diagnostic accuracy improved to 0.85 for the training set and 0.87 for the validation set when lesion location information was implemented. Diagnostic accuracy for predicting 3 molecular subtypes of grade II/III gliomas was 0.74 for the training set and 0.56 for the validation set with lesion location information implemented. Conventional MRI-based radiomics is one of the most promising strategies that may lead to a non-invasive diagnostic technique for molecular characterization of grade II/III gliomas.