Visual Classification System Research Articles

Reliability and Cross-Cultural Validation of the Turkish Version of the Visual Function Classification System (VFCS) for Children with Cerebral Palsy

ABSTRACT Purpose This correlational study aimed to determine the reliability and validity of the Turkish version of the Visual Function Classification System (VFCS) for children with CP. Method Two physiotherapists and caregivers (n = 56) applied the VFCS (n = 120) two times at a 15-day interval. Physiotherapists’ classifications were compared for interrater reliability. Same physiotherapists and caregivers’ classifications’ were compared for intrarater reliability. Expanded & Revised Gross Motor Function Classification System (GMFCS E&R), Manual Ability Classification System (MACS), and Communication Function Classification System (CFCS) were used for assessing validity of the culturally adapted VFCS. Results The interrater reliability between physiotherapists was excellent (ICC = 0.974, ICC = 0.988), while interrater reliability between the first physiotherapist and caregivers was good (ICC = 0.866, ICC = 0.893). The intrarater reliability was excellent (ICC = 0.971, ICC = 0.996, ICC = 0.984). A correlation was detected between the VFCS and the GMFCS E&R, MACS, and CFCS levels (r = 0.415, p < .001, r = 0.450, p < .001, r = 0.491, p < .001). Conclusion The findings of this study demonstrate the utility, compatibility and reliability of the Turkish version of the VFCS. Trial registration number NCT05102955

Functional classification systems in Brazilian children with cerebral palsy: Reliability and associations between functional levels

ABSTRACT Children with Cerebral palsy (CP) present movement and posture disorders. The Gross Motor Function Classification System (GMFCS), Manual Ability Classification System (MACS), Communication Function Classification System (CFCS), Eating and Drinking Ability Classification System (EDACS), and Visual Function Classification System (VFCS) enhance the understanding of their performance. We verified inter-rater reliability and associations between the classification levels. Physical therapists classified 100 Brazilian children with CP (3–17 years) according to GMFCS, MACS, CFCS, EDACS, and VFCS. To evaluate inter-rater reliability (Intraclass Correlation Coefficient-ICC) two independent examiners concurrently assessed a subset of 60 participants. According to Spearman’s correlation coefficients, there were associations between GMFCS/MACS (r = 0.81), GMFCS/CFCS (r = 0.70), MACS/CFCS (r = 0.73), GMFCS/VFCS (r = 0.61), MACS/VFCS (r = 0.61), CFCS/EDACS (r = 0.58), CFCS/VFCS (r = 0.50), and EDACS/VFCS (r = 0.45) (p < .05). The inter-rater reliability ranged from excellent (ICC = 0.93-0.99) to good (ICC = 0.89), p < .05. The classification systems are reliable, and the levels associated with each other in Brazilian children, especially the GMFCS, MACS, and CFCS.

Long-term developmental outcomes of children with congenital Zika syndrome.

Eight years after the epidemics in Brazil, children with congenital Zika syndrome (CZS) and their families confront ongoing health challenges. This study aims to characterize how virally induced prenatal brain injury impacts development and functional outcomes among children diagnosed with CZS. We performed a cross-sectional study of a consecutive series of children diagnosed with CZS. Using validated neurodevelopmental assessments, we evaluated gross motor function, manual ability, communication, eating and drinking, and visual function. Sixty children (29 males, and 31 females) met the inclusion criteria for the study. Comorbidities such as epilepsy (90.0%) and undernutrition (38.3%), along with clinical conditions including dysphagia (68.3%) and dependence on tube feeding (31.7%), were observed. Our results demonstrate a majority of children at level V - the most severe level within a five-tier system - in the Gross Motor Function (86.7%), Manual Ability (85.0%), Communication Function (68.3%), Eating and Drinking Ability (40.0%) Classification Systems, and level IV in the Visual Function Classification System (38.3%). CZS is associated with severe functional impairments and comorbidities, adversely impacting child development and quality of life. These findings reveal persistent challenges affecting the functioning of children with CZS, underscoring the need for continued support and specialized care. This study aimed to characterize the long-term clinical and functional characteristics of a subset of children with Congenital Zika Syndrome (CZS). We found that eight years after the Brazilian Zika epidemic, this subset of children with CZS continues to demonstrate major functional limitations impacting mobility, vision, and the ability to eat and drink. Our analysis documented a very high level of disability in several key functional classification systems. Notably, applying a new instrument for visual ability among children diagnosed with cerebral palsy, we found that more than 60% of the study group have poor or very poor visual function.

Multi-site Garbage Collection Robot based on Machine Vision

At present, the problem of garbage disposal has become an urgent problem to be solved in the global energy technology revolution and transformation and development. The mainstream garbage removal methods are manual picking and mechanical collection, which has low cleaning efficiency and high cleaning cost. In order to realize the convenient and efficient recycling and cleaning of land and water garbage, a new propeller dynamics model is proposed. Through the structural strength analysis and gas-liquid two-phase flow simulation analysis, combined with the aerodynamic theory, the propeller impeller model was optimized and its suitable working parameters were explored. Based on the paddle impeller mechanism, combined with the visual recognition and classification system controlled by the algorithm combining YOLO and template matching, a multi-site garbage recycling robot based on machine vision was designed, which realized the intelligent and multi-step integration of garbage recycling, and provided a new research idea for improving the way of garbage cleaning.

A Robust Study of High-redshift Galaxies: Unsupervised Machine Learning for Characterizing Morphology with JWST up to z ∼ 8

Galaxy morphologies provide valuable insights into their formation processes, tracing the spatial distribution of ongoing star formation and encoding signatures of dynamical interactions. While such information has been extensively investigated at low redshift, it is crucial to develop a robust system for characterizing galaxy morphologies at earlier cosmic epochs. Relying solely on nomenclature established for low-redshift galaxies risks introducing biases that hinder our understanding of this new regime. In this paper, we employ variational autoencoders to perform feature extraction on galaxies at z > 2 using JWST/NIRCam data. Our sample comprises 6869 galaxies at z > 2, including 255 galaxies at z > 5, which have been detected in both the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey Hubble Space Telescope fields and the Cosmic Evolution Early Release Science Survey done with JWST, ensuring reliable measurements of redshift, mass, and star formation rates. To address potential biases, we eliminate galaxy orientation and background sources prior to encoding the galaxy features, thereby constructing a physically meaningful feature space. We identify 11 distinct morphological classes that exhibit clear separation in various structural parameters, such as the concentration, asymmetry, and smoothness (CAS) metric and M 20, Sérsic indices, specific star formation rates, and axis ratios. We observe a decline in the presence of spheroidal-type galaxies with increasing redshift, indicating the dominance of disk-like galaxies in the early Universe. We demonstrate that conventional visual classification systems are inadequate for high-redshift morphology classification and advocate the need for a more detailed and refined classification scheme. Leveraging machine-extracted features, we propose a solution to this challenge and illustrate how our extracted clusters align with measured parameters, offering greater physical relevance compared to traditional methods.

A Visual Approach to the Damage Diagnosis of Monoliths from the Monquira Archaeological Park - El Infiernito in Colombia

Natural stones of monuments represent an important part of our world’s cultural heritage but deteriorate by the direct or indirect interaction of mineral and stone weathering. Application of damage indices improves stone damage diagnosis and is very appropriate for the determination of necessary preservation procedures as well as the long-term survey and maintenance of stone monuments. The sandstones used at El Infiernito in Villa de Leyva, Colombia, today show weathering damages of partly extreme intensity. In-situ surveys allowed damage mapping to identify major causes of damage and to understand their relationship with stone and environment characteristics. In this research, the results of describing the state of deterioration of the pre-Hispanic phallic monoliths in the Archaeological Park of Monquirá "El Infiernito", are presented in the form of monument visual posters. On the basis of the visualizing of weathering forms at representative monoliths, the state of deterioration is precisely characterized and evaluated. The monument mapping and photographic documentation method can also be useful for reassessing weather damage in the framework of long-term monitoring and maintenance of monuments. This article presents the methodology of mapping weathering forms of stone monuments by using a visual classification system developed over the years with various applications.

Development of an integrated telerehabilitation model in children with cerebral visual impairment in South India during the COVID-19 pandemic - A pilot study.

Coronavirus disease 2019 (COVID-19) pandemic affected the in-person rehabilitation/habilitation services in families with children with cerebral visual impairment (CVI) in India. This study aimed to develop a structured and family-centered telerehabilitation model alongside conventional in-person intervention in children with CVI to observe its feasibility in the Indian population. This pilot study included 22 participants with a median age of 2.5 years (range: 1-6) who underwent a detailed comprehensive eye examination followed by functional vision assessment. The visual function classification system (VFCS) was administered to the children and the structured clinical question inventory (SCQI) to the parents. Every participant underwent 3 months of telerehabilitation including planning, training, and monitoring by experts. At 1 month, the parental care and ability (PCA) rubric was administered to the parents. After 3 months, in an in-person follow-up, all the measures were reassessed for 15 children. After 3 months of Tele-rehabilitation there were significant improvements noted in PCA rubric scores (P<0.05). Also, statistically significant improvements were noted in functional vision measured using SCQI and VFCS scores (P<0.05) compared to baseline. The outcomes of the study provide the first steps towards understanding the use of a novel tele-rehabilitation model in childhood CVI along-side conventional face-to-face intervention. The added role of parental involvement in such a model is highly essential.

Digital Game Visual Style Classification: Influence of Information Quality Components on the User Satisfaction during Game Searching Activity in the Digital Game Library

The digital gaming community appreciates the visual style classification system to search for information about a game product. However, scholars have discovered that the applied visual style classification system frequently leads to frustration and dissatisfaction among users owing to inaccurate search results. Moreover, there are yet any research studies that accurately measure user satisfaction during game searching activity in the digital game library based on the current visual style classification system. Therefore, this study was performed to investigate the influence of information quality components (Accuracy, Content, Ease of Use, Format, and Timeliness) on User Satisfaction during game searching activity in the digital game library. A cross-sectional study was conducted by distributing self-administered digital questionnaires to 239 game players in Malaysia. The 12-item instrument survey questionnaire, which consists of four main sections, adopted the End-user Computing Satisfaction (EUCS) model to examine the relationships between user satisfaction and information quality components. The results were then analysed using descriptive statistics, preliminary data analysis, and Confirmatory Factor Analysis (CFA). Based on the findings, more than half of the respondents were male (n = 139, 58.2%) with an age range between 19 and 24 years old, and 79.1% (n = 189) have experience in searching digital games based on visual styles. The EUCS model revealed that user satisfaction during game searching activity using the visual style information highly correlated with Content, Accuracy, and Ease of Use. Users were satisfied when they receive accurate, precise, and sufficient information. In addition, a user-friendly and simplified navigation interface improved the searching experience and stimulate users to further their searching activity. In contrast, the Format and Timeliness showed a weak correlation. Providing visual classification and appeal format has less impact on user satisfaction. In addition, the fast speed of information retrieval and up-to-date information showed an insignificant contribution to user satisfaction. Overall, this research demonstrated that information quality components, namely Content, Accuracy, and Ease of Use, influenced User Satisfaction when searching for games based on the visual style. Future studies should explore and evaluate the effectiveness of visual style information systems for digital game distribution platforms in Malaysia.

Psychometric properties of the Viking Speech Scale-Turkish version for children with cerebral palsy aged 4-18 years based on live and video-based observation.

Speech is the most common method of communication. Video-based clinical communication evaluation is a requirement for children with speech-language impairments living in rural areas, and those who have limited mobility. To determine the validity and reliability of the Turkish version of the Viking Speech Scale (VSS-T) via live and video-based observation for children with cerebral palsy (CP) aged 4-18 years. A total of 142 children (mean age 8.18 ± 3.98 years; 68 female) with CP were included in this study. Their motor, communication, visual and eating-drinking function levels and comorbidities (dental, swallowing, cognitive impairments and epilepsy) were recorded. The Intelligibility in Context Scale (ICS), the Pediatric Evaluation of Disability Inventory-Social Function (PEDI-SF), and the Functional Independence Measure for Children-Communication (WeeFIM-C) were assessed to examine the concurrent validity of the VSS-T. The interrater reliability of the VSS-T was analysed between parents, physical therapists, and speech and language therapists from live and video-based observation. Intra-rater reliability was calculated from ratings made from live and video-based observations taken 3 weeks apart. The VSS-T was strongly related to the ICS (r = -0.830), PEDI-SF (r = -0.819), WeeFIM-C (r = -0.643), other functional classifications (r > 0.432), and the comorbidities (Cramer's V > 0.284, p < 0.001). Good to excellent interrater reliability (κw ≥ 0.838) and intra-rater reliability (intraclass correlation coefficient (ICC) = 0.848-0.995) were found between parents and therapists. Speech and language therapists, physical therapists, and parents can use the VSS-T as a valid and reliable classification system to describe speech intelligibility of 4-18-year-old children with CP. Both live and video-based observations can be used to administer the VSS-T. What is already known on the subject The English version of the VSS has been shown to be a valid and reliable tool used to classify the speech of children with CP aged 4-13 years. The scale can be administered by means of live observation of the child or based on clinicians' notes on the case by parents, SLTs, physiotherapists and paediatricians. What this paper adds to existing knowledge The VSS-T is valid and reliable for children with CP aged 4-18 years. Video-based observation is a suitable method for evaluating the VSS-T levels. The VSS-T has a moderate association with the CFCS. What are the potential or actual clinical implications of this work? The VSS-T is a valid and reliable method of categorizing the severity of motor speech impairment for Turkish children with CP in clinical research studies, registry systems or epidemiological studies. Both experienced and inexperienced therapists can use either live or video-based observation methods to administer the VSS-T. This study extended the validity and reliability of the scale in children with CP aged up to 18 years. The VSS-T is also associated with the Visual Functional Classification System (VFCS), which has been recently developed for describing the visual abilities of children with CP in daily life. In addition, the VSS-T is associated with the presence of dental, swallowing, cognitive problems and epilepsy.

Embedded Real-Time Clothing Classifier Using One-Stage Methods for Saving Energy in Thermostats

Energy-saving is a mandatory research topic since the growing population demands additional energy yearly. Moreover, climate change requires more attention to reduce the impact of generating more CO2. As a result, some new research areas need to be explored to create innovative energy-saving alternatives in electrical devices that have high energy consumption. One research area of interest is the computer visual classification for reducing energy consumption and keeping thermal comfort in thermostats. Usually, connected thermostats obrtain information from sensors for detecting persons and scheduling autonomous operations to save energy. However, there is a lack of knowledge of how computer vision can be deployed in embedded digital systems to analyze clothing insulation in connected thermostats to reduce energy consumption and keep thermal comfort. The clothing classification algorithm embedded in a digital system for saving energy could be a companion device in connected thermostats to obtain the clothing insulation. Currently, there is no connected thermostat in the market using complementary computer visual classification systems to analyze the clothing insulation factor. Hence, this proposal aims to develop and evaluate an embedded real-time clothing classifier that could help to improve the efficiency of heating and ventilation air conditioning systems in homes or buildings. This paper compares six different one-stage object detection and classification algorithms trained with a small custom dataset in two embedded systems and a personal computer to compare the models. In addition, the paper describes how the classifier could interact with the thermostat to tune the temperature set point to save energy and keep thermal comfort. The results confirm that the proposed real-time clothing classifier could be implemented as a companion device in connected thermostats to provide additional information to end-users about making decisions on saving energy.

An analysis of the appearance characteristics of Korean ceramics per era through statistical analysis of metadata annotated with a visual element classification system of ceramics

This study is to create metadata with a ceramic visual element classification system, such as color, form, material, and pattern, and then to analyze the visual characteristics of Korean ceramics by era through this metadata statistical analysis. To achieve this, first, the visual element classification system for ceramics was established. Second, 7,346 ceramic photos were acquired and annotated using the visual element classification system to create metadata. Third, through statistical analysis of the metadata, the preferred visual elements in each era were organized, and their characteristics were analyzed. In particular, the differences in form implementation, color technology, and pattern representation, which vary depending on material properties, were identified. Through this study, the flow of visual elements of Korean ceramics and the reason for each type of appearance and production method could be comprehended more systematically.

Visual or visual-tactile examination to detect and inform the diagnosis of enamel caries.

Visual or visual-tactile examination to detect and inform the diagnosis of enamel caries.

Beyond the hubble sequence – exploring galaxy morphology with unsupervised machine learning

ABSTRACT We explore unsupervised machine learning for galaxy morphology analyses using a combination of feature extraction with a vector-quantized variational autoencoder (VQ-VAE) and hierarchical clustering (HC). We propose a new methodology that includes: (1) consideration of the clustering performance simultaneously when learning features from images; (2) allowing for various distance thresholds within the HC algorithm; (3) using the galaxy orientation to determine the number of clusters. This set-up provides 27 clusters created with this unsupervised learning that we show are well separated based on galaxy shape and structure (e.g. Sérsic index, concentration, asymmetry, Gini coefficient). These resulting clusters also correlate well with physical properties such as the colour–magnitude diagram, and span the range of scaling relations such as mass versus size amongst the different machine-defined clusters. When we merge these multiple clusters into two large preliminary clusters to provide a binary classification, an accuracy of $\sim 87{{\ \rm per\ cent}}$ is reached using an imbalanced data set, matching real galaxy distributions, which includes 22.7 per cent early-type galaxies and 77.3 per cent late-type galaxies. Comparing the given clusters with classic Hubble types (ellipticals, lenticulars, early spirals, late spirals, and irregulars), we show that there is an intrinsic vagueness in visual classification systems, in particular galaxies with transitional features such as lenticulars and early spirals. Based on this, the main result in this work is not how well our unsupervised method matches visual classifications and physical properties, but that the method provides an independent classification that may be more physically meaningful than any visually based ones.

Smart Culture Lens: An Application That Analyzes the Visual Elements of Ceramics

The Smart Culture Lens developed in this study is an application developed by utilizing the visual element classification system of ceramic and AI technology. The user can analyze the visual elements of the ceramic photo taken with a smartphone and search for similar ceramics related to each visual element. For this development, as a first step, visual elements such as color, form, material, and pattern were defined as criteria for classifying ceramic appearance, and the visual element classification system of ceramics was organized. In the second step, 19,610 images of 7,346 ceramics were collected through museum visit photography and web, and a database was built by annotating these images with a visual element classification system. In the third step, representative object detection models, Faster R-CNN and Mask R-CNN were trained based on a ceramic classification system. Through those trained object detection models, visual elements and masks of the input image were recognized, and representative colors of the area were extracted using the k-means algorithm through the recognized masks. The performance of the trained object detection models (Average precision of Form / Material / Pattern 1st-level category = 0.87 / 0.89 / 0.72) shows that the amount of collected data and the established classification system are useful. Finally, by applying the above development results, a mobile application called 'Smart Culture Lens' was developed, and the usefulness of this application was confirmed through a user experience test. This study combines AI technology into cultural heritage so that people can intuitively explore artifacts from a new perspective, which differs from traditional artifact exploring methods. All the detailed processes of this development will be a guide to how to apply AI technology to the cultural heritage.

EQUSUM: Endometriosis QUality and grading instrument for SUrgical performance: proof of concept study for automatic digital registration and classification scoring for r-ASRM, EFI and Enzian.

STUDY QUESTIONIs electronic digital classification/staging of endometriosis by the EQUSUM application more accurate in calculating the scores/stages and is it easier to use compared to non-digital classification?SUMMARY ANSWERWe developed the first digital visual classification system in endometriosis (EQUSUM). This merges the three currently most frequently used separate endometriosis classification/scoring systems (i.e. revised American Society for Reproductive Medicine (rASRM), Enzian and Endometriosis Fertility Index (EFI)) to allow uniform and adequate classification and registration, which is easy to use. The EQUSUM showed significant improvement in correctly classifying/scoring endometriosis and is more user-friendly compared to non-digital classification.WHAT IS KNOWN ALREADYEndometriosis classification is complex and until better classification systems are developed and validated, ideally all women with endometriosis undergoing surgery should have a correct rASRM score and stage, while women with deep endometriosis (DE) should have an Enzian classification and if there is a fertility wish, the EFI score should be calculated.STUDY DESIGN, SIZE, DURATIONA prospective endometriosis classification proof of concept study under experts in deep endometriosis was conducted. A comparison was made between currently used non-digital classification formats for endometriosis versus a newly developed digital classification application (EQUSUM).PARTICIPANTS/MATERIALS, SETTING, METHODSA hypothetical operative endometriosis case was created and summarized in both non-digital and digital form. During European endometriosis expert meetings, 45 DE experts were randomly assigned to the classic group versus the digital group to provide a proper classification of this DE case. Each expert was asked to provide the rASRM score and stage, Enzian and EFI score. Twenty classic forms and 20 digital forms were analysed. Questions about the user-friendliness (system usability scale (SUS) and subjective mental effort questionnaire (SMEQ)) of both systems were collected.MAIN RESULTS AND THE ROLE OF CHANCEThe rASRM stage was scored completely correctly by 10% of the experts in the classic group compared to 75% in the EQUSUM group (P < 0. 01). The rASRM numerical score was calculated correctly by none of the experts in the classic group compared with 70% in the EQUSUM group (P < 0.01). The Enzian score was correct in 60% of the classic group compared to 90% in the EQUSUM group (P = 0.03). EFI scores were calculated correctly in 25% of the classic group versus 85% in the EQUSUM group (P < 0.01). Finally, the usability measured with the SUS was significantly better in the EQUSUM group compared to the classic group: 80.8 ± 11.4 and 61.3 ± 20.5 (P < 0.01). Also the mental effort measured with the SMEQ was significant lower in the EQUSUM group compared to the classic group: 52.1 ± 18.7 and 71.0 ± 29.1 (P = 0.04). Future research should further develop and confirm these initial findings by conducting similar studies with larger study groups, to limit the possible role of chance.LIMITATIONS, REASONS FOR CAUTIONThese first results are promising, however it is important to note that this is a preliminary result of experts in DE and needs further testing in daily practice with different types (complex and easy) of endometriosis cases and less experienced gynaecologists in endometriosis surgery.WIDER IMPLICATIONS OF THE FINDINGSThis is the first time that the rASRM, Enzian and EFI are combined in one web-based application to simplify correct and automatic endometriosis classification/scoring and surgical registration through infographics. Collection of standardized data with the EQUSUM could improve endometriosis reporting and increase the uniformity of scientific output. However, this requires a broad implementation.STUDY FUNDING/COMPETING INTEREST(S)To launch the EQUSUM application, a one-time financial support was provided by Medtronic to cover the implementation cost. No competing interests were declared.TRIAL REGISTRATION NUMBERN/A.

Prevalence and risk factors of perirenal hyperintensities on T2-weighted MR imaging.

To assess the prevalence and severity of perirenal hyperintensities (PRHs) on T2-weighted magnetic resonance imaging (MRI) and their risk factors in the general population. 1752 participants (910 women, 842 men, median age: 52years) of the Study of Health in Pomerania were included. A visual classification system was established to assess the severity of PRHs (stage 0: no PRHs visible to stage 3: extensive PRHs visible). The MRI-derived thickness of the perirenal and posterior pararenal space and additional risk factors were determined and associated with PRH stages. The overall prevalence of PRHs was 40.7% and was higher in men than in women. PRHs were more common around the left compared to the right kidney (p < 0.001). On both sides, the presence of PRHs was associated with age (p < 0.001). Multivariable-adjusted regression models revealed male sex (OR = 13.2, 95%CI 9.7; 17.9, p < 0.001), older age (OR = 1.1, 95%CI 1.1; 1.1, p < 0.001), higher body mass index (BMI; OR = 1.1, 95%CI 1.0; 1.1, p < 0.001), current smoking status (OR = 1.9, 95%CI 1.4; 2.6, p < 0.001), and greater thickness of the posterior pararenal space (OR = 1.2, 95%CI 1.2; 1.2, p < 0.001) as independent significant risk factors of left-sided PRHs. The same risk factors were confirmed for right PRHs. Glomerular filtration rate was not associated with PRHs on either sides (OR = 1.00, 95%CI 1.0;1.0, p > 0.05). PRHs are common in the general population, with a higher prevalence in men, elderly individuals, and around the left kidney. Furthermore, BMI, posterior pararenal space thickness, and current smoking, but not renal function itself, are associated with the presence of PRHs.

Value of Endolymphatic Hydrops and Perilymph Signal Intensity in Suspected Ménière Disease.

Ménière disease is characterized by endolymphatic hydrops, whereas perilymphatic enhancement on MR imaging has been suggested to be of additional value in diagnosing Ménière disease. This study evaluates the presence of endolymphatic hydrops and perilymphatic enhancement in patients with Ménière disease and with other vertigo-associated inner ear pathology. A 3D-FLAIR sequence 4 hours after intravenous gadolinium injection was performed to visualize the endolymph and perilymph in 220 patients suspected of having Ménière disease. Patients' ears were retrospectively categorized as having Ménière disease (probable or definite) or other vertigo-associated inner ear pathology not attributable to Ménière disease. Endolymphatic hydrops was evaluated using a visual classification system, and perilymphatic enhancement was scored both visually and quantitatively. Endolymphatic hydrops was present in 137 (91.9%) of the definite Ménière disease ears and in 9 (7.0%) of the ears with other vertigo-associated inner ear pathology (P < .001). The combination of endolymphatic hydrops and visually increased perilymphatic enhancement was present in 122 (81.9%) definite Ménière disease ears compared with 4 (3.1%) ears with other vertigo-associated inner ear pathology (P < .001). This combination increases the positive predictive value from 0.94 for endolymphatic hydrops and 0.91 for perilymphatic enhancement to 0.97. The addition of measured perilymphatic enhancement leads to a moderate decrease in sensitivity from 0.92 for endolymphatic hydrops to 0.86. The combination of perilymphatic enhancement and endolymphatic hydrops in patients suspected of having Ménière disease increases the positive predictive value in the diagnosis of definite Ménière disease.

A Survey on Large-Scale Machine Learning

Machine learning can provide deep insights into data, allowing machines to make high-quality predictions and having been widely used in real-world applications, such as text mining, visual classification, and recommender systems. However, most sophisticated machine learning approaches suffer from huge time costs when operating on large-scale data. This issue calls for the need of {Large-scale Machine Learning} (LML), which aims to learn patterns from big data with comparable performance efficiently. In this paper, we offer a systematic survey on existing LML methods to provide a blueprint for the future developments of this area. We first divide these LML methods according to the ways of improving the scalability: 1) model simplification on computational complexities, 2) optimization approximation on computational efficiency, and 3) computation parallelism on computational capabilities. Then we categorize the methods in each perspective according to their targeted scenarios and introduce representative methods in line with intrinsic strategies. Lastly, we analyze their limitations and discuss potential directions as well as open issues that are promising to address in the future.

Image Spam Filtering using Machine Learning Techniques

Unsolicited visual data is undesirable in any form. The art of hiding malicious content in images and adding them as attachments to electronic mails has become a popular nuisance. In recent years, attackers have developed various new techniques to evade traditional spam classification systems. Text-based spam classification has been in focus for a long time and, researchers have successfully created a prodigal system for identifying spam text in electronic mails using Optical Character Recognition technology. In the last decade, extensive work has been performed to tackle image spam but with unsatisfactory results. Various algorithms and data augmentation techniques are used today to develop an optimal model for image spam recognition. Many of these proposed systems come close to the ideal system but do not provide 100 percent accuracy. This paper highlights the role of three popular techniques in image spam filtering. We discuss the importance and application of Optical Character Recognition, Support Vector Machines and, Artificial Neural Networks in unsolicited visual data filtering. This paper sheds light on the algorithms of these techniques. We provide a comparison of their accuracy, which helps us draw useful insights for developing a robust unsolicited visual data classification system. This paper aims to bring clarity regarding the feasibility of using these techniques to develop an unsolicited visual data filtering system. This paper records that the most favourable results are obtained using Artificial Neural Networks.

Visual Function Classification System for children with cerebral palsy: development of a new tool

This commentary is on the original article by Baranello et al. on pages 104‐110 of this issue