Imaging Signs Research Articles

The Cobra sign: A marker for overestimation of tendon retraction in proximal semimembranosus tendon avulsion.

To introduce a new magnetic resonance imaging (MRI) sign, termed the Cobra sign, and identify its diagnostic metrics. The secondary aim was to demonstrate that this sign can be a source of false evaluation of tendon retraction in patients with proximal hamstring avulsion injury. This retrospective cohort study targeted patients surgically treated for proximal hamstring avulsion injury from January 2019 to June 2023. The MRI Cobra sign was defined as a wavy curved T2-hypointense band with the free end folding distally over itself, resembling a cobra head. The primary outcome measure was the characterization of the Cobra sign in patients with proximal hamstring avulsion injury. The secondary outcome was the association of this sign with tendon retraction. The study included 81 proximal hamstring avulsion injury patients (mean age of 45.7, SD = 13.9), with 41 (50.6%) complete avulsions, 33 semimembranosus, and 7 conjoint tendons. The MRI Cobra sign was found in 25 patients (17 semimembranosus and 8 complete). It was confirmed surgically only in semimembranosus cases. It demonstrated 51.5% sensitivity and 83.3% specificity for isolated semimembranosus avulsions, with a significant positive likelihood ratio of 3.0. MRI retraction was 10.05 cm (±3.0), reducing to 7.9 cm (±2.5) on surgical measurement (mean difference = 2.0 cm, p < 0.001). The regression analysis confirmed MRI retraction's influence on the Cobra sign, with a 1.4 odds increase per unit (p < 0.001). In linear regression analysis, each unit increase in MRI retraction corresponded to a 79% increase in surgical retraction (coefficient 0.7, t = 11.1, p < 0.001). The Cobra sign demonstrated acceptable diagnostic accuracy for isolated semimembranosus avulsion, with a high specificity of 83.3%, a low sensitivity of 51.5%, and a positive likelihood ratio of 3.0. The presence of the Cobra sign indicates an overestimated MRI retraction by approximately 21%. Level III.

Sign Language Detection Using Machine Learning

This paper investigates the application of machine learning for sign language detection. The objective is to develop a model that translates sign language into spoken language, bridging the communication gap between deaf and hearing individuals. You Only Look Once (YOLO), a deep learning object detection algorithm, is employed to train a model on a dataset of labeled sign language images derived from video data. The system achieves real-time sign detection in videos. However, challenges include the scarcity of large, labeled datasets and the inherent ambiguity of certain signs, which can lead to reduced detection accuracy. This research contributes to the field of Assistive Technologies (AT) by promoting accessibility and social inclusion for the deaf community.

A two-stage road sign detection and text recognition system based on YOLOv7

We developed a two-stage traffic sign recognition system to enhance safety and prevent tragic traffic incidents involving self-driving cars. In the first stage, YOLOv7 was employed as the detection model for identifying 31 types of traffic signs. Input images were set to 640 × 640 pixels to balance speed and accuracy, with high-definition images split into overlapping sub-images of the same size for training. The YOLOv7 model achieved a training accuracy of 99.2 % and demonstrated robustness across various scenes, earning a testing accuracy of 99 % in both YouTube and self-recorded driving videos. In the second stage, extracted road sign images underwent rectification before processing with OCR tools such as EasyOCR and PaddleOCR. Post-processing steps addressed potential confusion, particularly with city/town names. After extensive testing, the system achieved recognition rates of 97.5 % for alphabets and 99.4 % for Chinese characters. This system significantly enhances the ability of self-driving cars to detect and interpret traffic signs, thereby contributing to safer road travel.

Implementation of Sibi Sign Language Realtime Detection Program (Case Studi At Sekolah Luar Biasa Negeri 1 Tabanan)

Indonesian deaf people utilize SIBI to communicate using spoken words, gestures, facial expressions, and body language. SIBI, certified for Special Schools (SLB), helps deaf pupils communicate. This project implements SIBI (Indonesian Sign Language System) a real-time detection algorithm at Sekolah Luar Biasa Negeri 1 Tabanan using image processing and YoloV8 ultralytics deep learning. The program trains a sign language gesture detection model on Google Colab's GPU. The SIBI sign language images were used to train a YoloV8 object detection model. The camera captures movements, which the YoloV8 algorithm trained on SIBI gesture data processes. It can recognize gestures in real time and generate text to non-sign language users. The dataset has 107 class vocabulary and 7 class affix prefixes for complete gesture recognition. Shirt color, room brightness, and webcam quality affect detection rates. Optimal detection accuracy is 87.74% and subpar 58.02%. Despite these limitations, the strategy helps deaf students communicate more effectively with non-sign language speakers. This program improves inclusivity and communication in schools, making learning easier for hearing-impaired pupils. This work provides a reliable and quick sign language identification system to help deaf educators and caregivers with daily interactions and education.

Associations between semi-quantitative evaluation of intracranial arterial calcification and total cerebral small vessel disease burden score: a retrospective case-control study.

Arteriosclerotic cerebral small vessel disease (aCSVD) is a cause of cognitive impairment, dementia, and stroke. Developing a better understanding of the risk factor of aCSVD is key to reducing the incidence of these conditions. This study investigated the association between intracranial arterial calcification (IAC) and total cerebral small vessel disease (CSVD) burden score. This is a retrospective study, the subjects were transient ischemic attack (TIA) or acute ischemic stroke (AIS) patients. The data of 303 inpatients admitted to our study hospital between December 2018 and July 2020 were analyzed. Four imaging markers of CSVD (lacunes, white matter hyperintensities, cerebral microbleeds, and enlarged perivascular spaces) were evaluated by magnetic resonance imaging, and a total CSVD burden score was calculated. The experimental group was divided into four subgroups according to total CSVD burden score (1-4 points). Patients without CSVD (0 points) served as the control group. Head computerized tomography (CT) scans were used to assess ICA, using Babiarz's method. The correlations between IAC and single imaging markers of CSVD were determined using Spearman's rank correlation. Binary logic regression analysis and multivariate ordered logic regression analysis were used to determine the associations between IAC and aCSVD. IAC was positively correlated with total CSVD burden score (r = 0.681), deep white matter hyperintensities (r = 0.539), periventricular white matter hyperintensities (r = 0.570), cerebral microbleeds (r = 0.479), lacunes (r = 0.541), and enlarged perivascular spaces (r = 0.554) (all p < 0.001). After adjusting for the confounding factors of age, diabetes, and hypertension, aCSVD was independently associated with IAC grade 1-2 [odds ratio (OR) = 23.747, 95% confidence interval (CI) = 8.376-67.327] and IAC grade 3-4 (OR = 30.166, 95% CI = 8.295-109.701). aCSVD severity was independently associated with IAC grade 3-4 (OR = 4.697, 95% CI = 1.349-16.346). IAC is associated with the total CSVD burden score and single imaging signs.

The "Hypointense Focal Brain" on susceptibility-weighted imaging as a sign of venous congestion in cranial dural arteriovenous fistulas.

Cranial dural arteriovenous fistulas (dAVFs) are complex neurovascular malformations accounting for approximately 10%-15% of all intracranial arteriovenous malformations. The objective is to investigate the utility of susceptibility-weighted imaging (SWI) in identifying "hypointense focal brain" as an additional helpful sign of venous congestion in cranial dAVFs. A retrospective review of patients diagnosed with cranial dAVFs between January 2015 and June 2023 was conducted, and SWI was used to identify the "hypointense focal brain" sign within the venous drainage region of the dAVF. The "hypointense focal brain" on SWI was identified as a low-intensity signal within the venous drainage region, indicative of venous congestion. The presence of this imaging sign was assessed by two neuroradiologists and signal intensity measurements were performed to support the presence of the sign. The study included six patients with cranial dAVFs exhibiting cortical venous retrograde drainage and the "hypointense focal brain" on SWI. Follow-up imaging post-treatment revealed resolution or improvement of the hypointense signal, confirming its association with venous congestion. Signal intensity measurements further supported the presence of this imaging sign in pre-treatment scans. The study's findings demonstrate the presence of a reversible "hypointense focal brain" sign on SWI in patients with cranial dAVFs and CVR, which can be useful as an additional imaging sign for venous congestion.

CT radiomics combined with clinical and radiological factors predict hematoma expansion in hypertensive intracerebral hemorrhage.

This study aimed to establish a hematoma expansion (HE) prediction model for hypertensive intracerebral hemorrhage (HICH) patients by combining CT radiomics, clinical information, and conventional imaging signs. A retrospective continuous collection of HICH patients from three medical centers was divided into a training set (n = 555), a validation set (n = 239), and a test set (n = 77). Extract radiomics features from baseline CT plain scan images and combine them with clinical information and conventional imaging signs to construct radiomics models, clinical imaging sign models, and hybrid models, respectively. The models will be evaluated using the area under the curve (AUC), clinical decision curve analysis (DCA), net reclassification index (NRI), and integrated discrimination improvement (IDI). In the training, validation, and testing sets, the radiomics model predicts an AUC of HE of 0.885, 0.827, and 0.894, respectively, while the clinical imaging sign model predicts an AUC of HE of 0.759, 0.725, and 0.765, respectively. Glasgow coma scale score at admission, first CT hematoma volume, irregular hematoma shape, and radiomics score were used to construct a hybrid model, with AUCs of 0.901, 0.838, and 0.917, respectively. The DCA shows that the hybrid model had the highest net profit rate. Compared with the radiomics model and the clinical imaging sign model, the hybrid model showed an increase in NRI and IDI. The hybrid model based on CT radiomics combined with clinical and radiological factors can effectively individualize the evaluation of the risk of HE in patients with HICH. CT radiomics combined with clinical information and conventional imaging signs can identify HICH patients with a high risk of HE and provide a basis for clinical-targeted treatment. HE is an important prognostic factor in patients with HICH. The hybrid model predicted HE with training, validation, and test AUCs of 0.901, 0.838, and 0.917, respectively. This model provides a tool for a personalized clinical assessment of early HE risk.

The Optimization of Numerical Algorithm Parameters with a Genetic Algorithm to Animate Letters of the Sign Alphabet

At present, the development of animation-based works for human–computer interaction applications has increased. To generate animations, actions are pre-recorded and animation flows are configured. In this research, from two images of letters of the sign language alphabet, intermediate frames were generated using a numerical traced algorithm based on homotopy. The parameters of a homotopy curve were optimized with a genetic algorithm to generate intermediate frames. In the experiments performed, sequences where a person executes pairs of letters in sign language were recorded and animations of the same pairs of letters were generated with the proposed method. Subsequently, the similarity of the real sequences to the animations was measured using Dynamic Time Wrapping. The results obtained show that the images obtained are consistent with their execution by a person. Animation files between sign pairs were created from sign images, with each file weighing an average of 18.3 KB. By having sequences between pairs of letters it is possible to animate words and sentences. The animations generated by this homotopy-based animation method optimized with a genetic algorithm can be used in various deaf interaction applications to provide assistance. From several pairs of letters a file base was generated using the animations between pairs of letters; with these files you can create animations of words and sentences.

Computed tomography-based radiomics combined with machine learning allows differentiation between primary intestinal lymphoma and Crohn's disease.

Due to similar clinical manifestations and imaging signs, differential diagnosis of primary intestinal lymphoma (PIL) and Crohn's disease (CD) is a challenge in clinical practice. To investigate the ability of radiomics combined with machine learning methods to differentiate PIL from CD. We collected contrast-enhanced computed tomography (CECT) and clinical data from 120 patients form center 1. A total of 944 features were extracted single-phase images of CECT scans. Using the last absolute shrinkage and selection operator model, the best predictive radiographic features and clinical indications were screened. Data from 54 patients were collected at center 2 as an external validation set to verify the robustness of the model. The area under the receiver operating characteristic curve, accuracy, sensitivity and specificity were used for evaluation. A total of five machine learning models were built to distinguish PIL from CD. Based on the results from the test group, most models performed well with a large area under the curve (AUC) (> 0.850) and high accuracy (> 0.900). The combined clinical and radiomics model (AUC = 1.000, accuracy = 1.000) was the best model among all models. Based on machine learning, a model combining clinical data with radiologic features was constructed that can effectively differentiate PIL from CD.

Magnetic Resonance Imaging Signs of Idiopathic Intracranial Hypertension

The magnetic resonance imaging (MRI) criteria currently used to diagnose idiopathic intracranial hypertension (IIH) are based on expert opinion and have limited accuracy. Additional neuroimaging signs have been proposed and used with contradictory results; thus, prospective evidence is needed to improve diagnostic accuracy. To provide evidence-based, accurate MRI signs for IIH diagnosis. This prospective cohort study was conducted from January 2018 to May 2021 with 3 validation cohorts at 2 Danish headache centers and with 3 independent international cohorts. Consecutive patients with suspected IIH were enrolled. Eligibility required the clinical suspicion of IIH, age 18 years or older, and written informed consent. The validation cohorts comprised patients with confirmed IIH from Austria and the US, and patients without IIH from the US. Data analysis was performed from December 2021 to August 2023. Standardized diagnostic workup was performed to classify cases according to current criteria, and blinded evaluation of cerebral radiological diagnostics and papilledema was performed. The primary outcomes were MRI signs associated with IIH as assessed by univariate analyses. An MRI score estimating papilledema was calculated using machine learning. Internal validation of associations with lumbar puncture opening pressure and outcome and external validation of accuracy were performed in 3 cohorts. Of 192 eligible patients (185 women [96.4%]; median [IQR] age, 28.0 [23.0-35.0] years), 110 were classified as having IIH, 4 as having probable IIH, and 1 as having suspected IIH without papilledema; 77 did not have IIH and served as controls, with corresponding age, sex, and weight. Papilledema at diagnosis was associated with perioptic subarachnoid space distension (56 patients [68.3%] vs 21 patients [41.2%]), posterior globe flattening (53 patients [66.3%] vs 10 patients [21.3%]), optic nerve disc protrusion (35 patients [30.4%] vs 2 patients [2.3%]), and transverse sinus venous stenosis (75 patients [79.8%] vs 29 patients [46.8%]). The papilledema-estimating MRI score showed optimal balance between sensitivity (49%) and specificity (87%) when 2 of the 3 latter signs were present and was associated with the lumbar puncture opening pressure and ophthalmological outcome. The score showed strong diagnostic accuracy in the external validation cohorts (587 patients; area under the receiver operating characteristic curve, 0.86) and outperformed the current (2013) diagnostic MRI criteria. The findings of this cohort study of patients with IIH and controls suggest that an evidence-based MRI score including posterior globe flattening, optic nerve disc protrusion, and transverse sinus stenosis can estimate the presence of papilledema more accurately than the current diagnostic criteria.

Analysis of malignancy signs in breast magnetic resonance imaging

Analysis of malignancy signs in breast magnetic resonance imaging

Application of artificial intelligence-based magnetic resonance imaging in diagnosis of cerebral small vessel disease.

Cerebral small vessel disease (CSVD) is an important cause of stroke, cognitive impairment, and other diseases, and its early quantitative evaluation can significantly improve patient prognosis. Magnetic resonance imaging (MRI) is an important method to evaluate the occurrence, development, and severity of CSVD. However, the diagnostic process lacks quantitative evaluation criteria and is limited by experience, which may easily lead to missed diagnoses and misdiagnoses. With the development of artificial intelligence technology based on deep learning, the extraction of high-dimensional features in imaging can assist doctors in clinical decision-making, and it has been widely used in brain function and mental disorders, and cardiovascular and cerebrovascular diseases. This paper summarizes the global research results in recent years and briefly describes the application of deep learning in evaluating CSVD signs in MRI imaging, including recent small subcortical infarcts, lacunes of presumed vascular origin, vascular white matter hyperintensity, enlarged perivascular spaces, cerebral microbleeds, brain atrophy, cortical superficial siderosis, and cortical cerebral microinfarct.

TRAFFIC SIGN RECOGNITION IN CHALLENGING WEATHER CONDITIONS USING CONVOLUTIONAL NEURAL NETWORKS

The ability of autonomous driving systems to recognize traffic signs in a variety of environmental conditions is crucial to their reliability. This study uses convolutional neural networks (CNNs) to provide a novel method for improving the accuracy of traffic sign recognition systems under challenging weather situations. The research focuses on developing a CNN-based model that is trained using the augmented German Traffic Sign Recognition Benchmark (GTSRB) dataset, which contains over 50,000 labeled images of road signs across 43 categories. To attempt to overcome the limitations of traditional CNN models under adverse environmental conditions, this work presents adaptive feature extraction layers that are especially made to reduce visibility problems brought on by rain, fog, and snow. By taking a comprehensive approach, the model uses advanced data augmentation methods to simulate different weather scenarios, greatly increasing the diversity of the training dataset. Through an analysis of theoretical and practical aspects, the study demonstrates how CNNs enhance the accuracy and efficiency of road sign detection systems in a different weather condition. Research evaluates the model's effectiveness using metrics such as precision, recall, and F1-score, proving its capability to reduce false positives and accurately detect relevant road sign instances. Additionally, the paper highlights the importance of careful dataset preparation and augmentation, model optimization, and training enhancements to improve the performance of road sign detection systems. The results of this research offer benefits for intelligent transport systems, autonomous driving, and road safety, pointing towards further advancements in precise and dependable road sign recognition technology.

Exploring deep learning: Preventing HIV through social media data

This paper explores the potential of deep learning in analyzing social media data to identify and support populations at high risk for HIV. With the widespread use of social media, there is an opportunity to leverage this data source for public health research and intervention. Deep learning, a subset of machine learning that utilizes artificial neural networks to analyze complex data, offers a promising approach to extract meaningful insights from large volumes of social media data. The paper begins with an overview of the HIV epidemic and the importance of early detection and prevention efforts. It then introduces deep learning and its applications in public health, highlighting its ability to analyze unstructured data such as text, images, and videos. A literature review is conducted to examine previous studies on using social media data for health surveillance and the applications of deep learning in this context. The review discusses the challenges and limitations of using social media data for public health research, including issues related to privacy, data bias, and algorithm transparency. The methodology section outlines the data collection process, including the sources of social media data such as Twitter and Facebook, and the preprocessing steps to clean and prepare the data for analysis. The section also describes the deep learning models selected for analyzing social media content and the evaluation metrics used to assess their performance. Case studies are presented to illustrate the application of deep learning in identifying HIV-related discussions on social media platforms, analyzing images for signs of risky behavior, and tracking the spread of HIV-related rumors and misinformation. Ethical considerations related to privacy, data bias, and algorithm transparency are discussed, along with recommendations for future research and application. The paper concludes with a summary of key findings and a call to action for further exploration and implementation of deep learning in public health.

Awareness and Detection of the Gridlock Manifestation for Vehicles Voice Alert using CNN

To ensure a smooth and secure flow of traffic, road signs are essential. A major cause of road accidents is negligence in viewing the Traffic signboards. The proposed system helps in recognizing the traffic sign and sending a voice alert through the speaker to the driver so that he or she may take necessary decisions. The proposed system is trained using a convolutional neural network (CNN) which helps in traffic sign image recognition and classification. Following the detection of the sign by the system, a voice alert is sent through the speaker which notifies the driver and also contains a section where the vehicle driver is alerted about the traffic signs. The aim of this system is to ensure the safety of the vehicle’s driver and passengers. Keywords: convolutional neural network (CNN), deep learning, traffic signs.

Sign Detection Model for the Deaf People using Deep Learning

Sign detection is a crucial task in various computer vision applications, including autonomous driving, robotics, and augmented reality. This paper proposes a novel sign detection model based on deep learning techniques. The proposed model leverages convolutional neural networks (CNNs) to automatically learn discriminative features from sign images, enabling robust detection performance across diverse environmental conditions. Keywords: sign detection, deep learning, convolution neural networks, computer vision, autonomous driving, robotics, augmented reality.

Early detection of pancreatic cancer

The diagnosis of pancreatic cancer associates an appalling significance. Detection of preinvasive stage of pancreatic cancer will ameliorate the survival of this deadly disease. Premalignant lesions such as Intraductal Papillary Mucinous Neoplasms or Mucinous Cystic Neoplasms of the pancreas are detectable on imaging exams and this permits their management prior their invasive development. Pancreatic intraepithelial neoplasms (PanIN) are the most frequent precursors of pancreatic adenocarcinoma (PDAC), and its particular type PanIN high-grade represents the malignant non-invasive form of PDAC. Unfortunately, PanINs are not detectable on radiologic exams. Nevertheless, they can associate indirect imaging signs which would rise the diagnostic suspicion. When this suspicion is established, the patient will be enrolled in a follow-up strategy that includes performing of blood test and serial imaging test such as computed tomography or magnetic resonance imaging, which will cost in the best-case scenario a burden of healthcare systems, and potential mortality in the worst-case scenario when the patient underwent resection surgery, worthless when there is no moderate or high grade dysplasia in the final histopathology. This issue will be avoid having at its disposal a diagnostic technique capable of detecting high-grade PanIN lesions, such is the cytology of pancreatic juice obtained by nasopancreatic intubation. Herein, we review the possibility of detection of early malignant lesions before they become invasive PADC.

Melorheostosis (Leri's Disease): A Review.

Melorheostosis is a noncancerous bone disease characterized by abnormal bone and soft tissue growth. Despite being identified almost a century ago, there are still many unknown aspects surrounding this condition. It can often be an incidental discovery, with patients experiencing associated pain and deformities. Diagnosis typically relies on X-rays, although not all cases exhibit the classic candle wax appearance. A new imaging sign known as the "dumpling on a plate sign" has been proposed for flat bones for both MRI and CT scans. A biopsy may be necessary in cases of uncertainty, as there is not a definitive histological feature. It is not uncommon for melorheostosis to be linked with other conditions, and a collaborative approach involving a multidisciplinary team should be considered. This condition should be considered in the differential diagnosis of sclerotic bone conditions. Management is generally aimed at symptom relief, either through conservative measures or surgical intervention.

Classification of the relationship between suprasellar arachnoid cyst and hydrocephalus based on treatment modalities: shunting versus neuroendoscopic approaches.

Children diagnosed with suprasellar arachnoid cysts often concurrently have hydrocephalus. This study aims to classify the relationship between suprasellar arachnoid cysts and hydrocephalus, discussing surgical strategies-shunting or neuroendoscopic approaches-and their sequence, based on this classification. A retrospective analysis was conducted on 14 patients diagnosed with suprasellar arachnoid cysts and hydrocephalus, treated surgically by the first author between January 2016 and December 2020. Clinical features, radiological findings, surgical strategies, and outcomes were reviewed. The classification of the relationship between the suprasellar arachnoid cysts and hydrocephalus was developed and illustrated with specific cases. Recommendations for future surgical management based on this classification are provided. We classified the relationship between suprasellar arachnoid cysts and hydrocephalus into three categories. SACH-R1, the direct type, represents cases where the cysts cause obstructive hydrocephalus. Here, neuroendoscopic ventriculocystocisternostomy (VCC) effectively treats both conditions. SACH-R2, the juxtaposed type, involves concurrent occurrences of cysts and hydrocephalus without a causative link. This is further subdivided into SACH-R2a, where acute progressive communicating hydrocephalus coexists with the cyst, initially managed with a ventriculoperitoneal shunt, followed by VCC upon stabilization of hydrocephalus; and SACH-R2b, where the cyst coexists with chronic stable communicating hydrocephalus, first addressed with VCC, followed by monitoring and potential secondary shunting if needed. Key factors differentiating SACH-R2a from SACH-R2b include the patient's age, imaging signs of fourth ventricle and cisterna magna enlargement, and the rapid progression or chronic stability and severity of hydrocephalus symptoms. SACH-R3, the reverse type, describes scenarios where shunting for hydrocephalus leads to the development or enlargement of the cyst, managed via neuroendoscopic VCC with precautions to prevent infections in existing shunt systems. The simultaneous presence of suprasellar arachnoid cysts and hydrocephalus requires a nuanced understanding of their complex relationship for optimal surgical intervention. The analysis and classification of their relationship are crucial for determining appropriate surgical approaches, including the choice and sequence of shunting and neuroendoscopic techniques. Treatment should be tailored to the specific type identified, rather than blindly opting for neuroendoscopy. Particularly for SACH-R2a cases, we recommend initial ventriculoperitoneal shunting.

Traffic Sign Classification Using CNN

This study presents a novel approach for traffic sign classification leveraging Convolutional Neural Networks (CNNs). With the proliferation of autonomous vehicles and advanced driver assistance systems, accurate and efficient traffic sign recognition is imperative for safe and efficient navigation. The proposed CNN-based method utilizes deep learning techniques to automatically extract discriminative features from traffic sign images, enabling robust classification across diverse environmental conditions and variations in sign appearance. Experimental results demonstrate the effectiveness of the proposed approach. This research contributes to the advancement of intelligent transportation systems by providing a reliable and scalable solution for real-time traffic sign classification. Our findings highlight the potential of deep learning for robust and accurate traffic sign classification in real-world scenarios.