Knowledge Of Modalities Research Articles

Implicit Modality Knowledge Alignment and Uncertainty Estimation for visible-infrared person re-identification

Visible-infrared person re-identification (VI-ReID) is a crucial cross-modal matching task in computer vision. Existing research typically relies on a single auxiliary modality to address the semantic gap in cross-modal data, but this approach fails to fully exploit the information inherent in both modalities. To tackle this issue, we propose a novel framework called Implicit Modality Knowledge Alignment (IMKA) and Uncertainty Estimation (UE). This framework aims to enhance the robustness of the learned common embedding space and improve the prediction accuracy of the VI-ReID model. The IMKA module is designed to extract valuable supplementary information from the original modality data to generate implicit modality data. The SKA module then aligns the distribution of significant knowledge learned from this generated implicit modality data. Meanwhile, the UE strategy is implemented to mitigate the overconfidence in incorrect predictions. The uncertainty of predicted probabilities is evaluated in a Dirichlet space, and the designed evidence loss bolsters the confidence in uncertainty. This uncertain information can enhance retrieval performance. Extensive experimental results on two publicly available VI-ReID datasets demonstrate that our IMKA-UE framework significantly outperforms state-of-the-art methods. The code for our IMKA-UE framework is available at https://github.com/SWU-CS-MediaLab/IMKA-UE.

Parental Knowledge, Attitudes, and Utilization of Preventive Dentistry Modalities in Al Qassim, Saudi Arabia: A Cross-Sectional Study.

Background Parents play a vital role in modeling good oral hygiene practices for their children; thus, their attitude, knowledge, and awareness toward preventive dentistry can be one of the major factors contributing to children's oral health. Objectives The objective of this study is to determine the parental knowledge and attitudes toward the modalities of preventive dentistry for their children in Al Qassim, Saudi Arabia. Materials and methods A cross-sectional study was carried out using an online self-administered questionnaire designed for parents living in the Al Qassim region. The questionnaire included demographic data, parent knowledge and attitude toward preventive dentistry, and awareness of preventive dentistry procedures. The data collected were analyzed using RStudio. Results Data were collected from 392 Saudi parents (male: n = 190 (48.5%), female: n = 202 (51.5%)). The majority of participants (n = 336, 85.7%) considered the importance of primary teeth for general body health, while 320 participants (81.6%) believed that primary dentition influences permanent teeth. Awareness regarding special preventive measures like using toothpaste with fluorides comprised 87.5% (n = 343), and that of the benefits of early dental visits was 77.3% (n = 303). However, there was marked unawareness and use of preventive procedures such as serial extraction (n = 226, 57.7% of respondents unaware) and silver diamine fluoride (n = 276, 70.4% unaware). The higher levels of education and income were positively related to better knowledge and attitudes about preventive dentistry. Conclusions The current paper highlights the importance of educational interventions for enhancing parental knowledge and awareness about preventive dentistry. Community-based educational programs and increased availability of preventive dentistry services might be helpful in eliminating the knowledge gap among parents about good oral health practice.

Uni-to-Multi Modal Knowledge Distillation for Bidirectional LiDAR-Camera Semantic Segmentation.

Combining LiDAR points and images for robust semantic segmentation has shown great potential. However, the heterogeneity between the two modalities (e.g. the density, the field of view) poses challenges in establishing a bijective mapping between each point and pixel. This modality alignment problem introduces new challenges in network design and data processing for cross-modal methods. Specifically, 1) points that are projected outside the image planes; 2) the complexity of maintaining geometric consistency limits the deployment of many data augmentation techniques. To address these challenges, we propose a cross-modal knowledge imputation and transition approach. First, we introduce a bidirectional feature fusion strategy that imputes missing image features and performs cross-modal fusion simultaneously. This allows us to generate reliable predictions even when images are missing. Second, we propose a Uni-to-Multi modal Knowledge Distillation (U2MKD) framework, leveraging the transfer of informative features from a single-modality teacher to a cross-modality student. This overcomes the issues of augmentation misalignment and enables us to train the student effectively. Extensive experiments on the nuScenes, Waymo, and SemanticKITTI datasets demonstrate the effectiveness of our approach. Notably, our method achieves an 8.3 mIoU gain over the LiDAR-only baseline on the nuScenes validation set and achieves state-of-the-art performance on the three datasets.

Abilities and the Epistemology of Ordinary Modality

Over the past two decades, modal epistemology has turned its attention to ordinary modal knowledge. This paper brings to the fore a neglected but central form of ordinary modal knowledge: knowledge of agentive modality, and in particular of our own abilities, which I call ‘ability knowledge’. I argue that modal epistemology as it is does not account for ability knowledge, by looking at the most promising candidate theories: perception-based, counterfactual-based, and similarity-based modal epistemologies. I then outline a more promising epistemology for our ability knowledge, which relies on the experience of our own agency, and draw out some lessons for modal epistemology in general.

Acne vulgaris in black pediatric patients: clinical presentation, treatment patterns, and unique needs.

Acne vulgaris is a common dermatological diagnosis observed in pediatric patients with skin of color, often resulting in scarring, keloid formation, and post-inflammatory hyperpigmentation, significantly impacting their quality of life. This exploratory retrospective chart review included 77 black pediatric patients seen at a tertiary care center for acne vulgaris between 2018 and 2023. We analyzed demographics, acne descriptors, and treatment modalities. The most common acne morphology was comedonal acne (83.6%), with 71% of the patients being female. Significant age differences were observed particularly for acne at the chin and overall face. Treatment regimens commonly prescribed included combinations of adapalene and benzoyl peroxide (22%), topical antibiotics, tretinoin, and benzoyl peroxide (34%). Given the higher risk of sequelae for patients with darker skin, it is crucial to address their unique treatment needs. This study highlights the distinctive characteristics of acne in black pediatric patients and calls for further research to enhance our understanding and treatment of this population. Limitations include the lack of direct patient interactions and reliance on chart data. Further studies are needed to compare acne presentation in skin of color of other populations, refining our knowledge of acne clinical presentation, complications, and treatment modalities for diverse patient populations.

Modal Knowledge for Expressivists

What does ‘Smith knows that it might be raining’ mean? Expressivism here faces a challenge, as its basic forms entail a pernicious type of transparency, according to which ‘Smith knows that it might be raining’ is equivalent to ‘it is consistent with everything that Smith knows that it is raining’ or ‘Smith doesn’t know that it isn’t raining’. Pernicious transparency has direct counterexamples and undermines vanilla principles of epistemic logic, such as that knowledge entails true belief and that something can be true without one knowing it might be. I re-frame the challenge in precise terms and propose a novel expressivist formal semantics that meets it by exploiting (i) the topic-sensitivity and fragmentation of knowledge and belief states and (ii) the apparent context-sensitivity of epistemic modality. The resulting form of assertibility semantics advances the state of the art for state-based bilateral semantics by combining attitude reports with context-sensitive modal claims, while evading various objectionable features. In appendices, I compare the proposed system to Beddor and Goldstein’s ‘safety semantics’ and discuss its analysis of a modal Gettier case due to Moss.

An epistemic distinction among essences, its metaphysical ground, and the role of philosophy

Uniformism is the view that one and the same epistemology should apply for all modal knowledge. I argue that, whether or not all modal knowledge can be accounted for in terms of knowledge of essences, uniformism about knowledge of essences is untenable. I do this by showing that, while some essences are empirically discoverable, others are not. I then argue that the uniquely realisable–non-uniquely realisable distinction is a better metaphysical candidate for grounding this epistemic difference than the concrete–abstract distinction. I conclude by suggesting that the task of philosophy is to identify non-empirically discoverable essences.

Response Evaluation Criteria in Gastrointestinal and Abdominal Cancers: Which to Use and How to Measure.

As the management of gastrointestinal malignancy has evolved, tumor response assessment has expanded from size-based assessments to those that include tumor enhancement, in addition to functional data such as those derived from PET and diffusion-weighted imaging. Accurate interpretation of tumor response therefore requires knowledge of imaging modalities used in gastrointestinal malignancy, anticancer therapies, and tumor biology. Targeted therapies such as immunotherapy pose additional considerations due to unique imaging response patterns and drug toxicity; as a consequence, immunotherapy response criteria have been developed. Some gastrointestinal malignancies require assessment with tumor-specific criteria when assessing response, often to guide clinical management (such as watchful waiting in rectal cancer or suitability for surgery in pancreatic cancer). Moreover, anatomic measurements can underestimate therapeutic response when applied to molecular-targeted therapies or locoregional therapies in hypervascular malignancies such as hepatocellular carcinoma. In these cases, responding tumors may exhibit morphologic changes including cystic degeneration, necrosis, and hemorrhage, often without significant reduction in size. Awareness of pitfalls when interpreting gastrointestinal tumor response is required to correctly interpret response assessment imaging and guide appropriate oncologic management. Data-driven image analyses such as radiomics have been investigated in a variety of gastrointestinal tumors, such as identifying those more likely to respond to therapy or recur, with the aim of delivering precision medicine. Multimedia-enhanced radiology reports can facilitate communication of gastrointestinal tumor response by automatically embedding response categories, key data, and representative images. ©RSNA, 2024 Test Your Knowledge questions for this article are available in the supplemental material.

EEG-eye movement based subject dependence, cross-subject, and cross-session emotion recognition with multidimensional homogeneous encoding space alignment

The joint learning of multimodal is helpful to extract the general information cross-modality in improving the performance of multimodal emotion recognition. However, focusing on a single common pattern can cause multimodal data to deviate from its original distribution and fail to fully capture the potential representation of the data. Therefore, we propose a multi-dimensional homogenous encoding spatial alignment (MHESA) method, which consists of two parts: multi-modal joint learning and modal knowledge transfer. To obtain a common projection space of EEG-EM features, we use a multimodal joint space encoder to learn the homogeneous joint space of EEG-Eye Movement (EM). To obtain a homogeneous encoding space based on modal knowledge, the knowledge transfer module learns the spatial distribution of EM features while retaining the original EEG features. The output of each module is used to construct a multidimensional homogeneous encoding space. The weight and multi-task loss function of the multi-dimensional homogeneous encoding space are dynamically adjusted by the Multi-task Joint Optimization Strategy (MJOS). By analyzing the effect of multi-task optimization, we found that compared with the subject dependence scene, the cross-subject scene has an advantage in the construction of joint encoding space, and the modal knowledge transfer feature has a higher contribution degree in cross-session. The experimental results show the MHESA method can make the model achieve more stable performance in three emotion recognition scenes.

Chemical species recognition in an adaptive radiation of Hawaiian Tetragnatha spiders (Araneae: Tetragnathidae).

Studies of adaptive radiations have played a central role in our understanding of reproductive isolation. Yet the focus has been on human-biased visual and auditory signals, leaving gaps in our knowledge of other modalities. To date, studies on chemical signals in adaptive radiations have focused on systems with multimodal signalling, making it difficult to isolate the role chemicals play in reproductive isolation. In this study we examine the use of chemical signals in the species recognition and adaptive radiation of Hawaiian Tetragnatha spiders by focusing on entire communities of co-occurring species, and conducting behavioural assays in conjunction with chemical analysis of their silks using gas chromatography-mass spectrometry. Male spiders significantly preferred the silk extracts of conspecific mates over those of sympatric heterospecifics. The compounds found in the silk extracts, long chain alkyl methyl ethers, were remarkably species-specific in the combination and quantity. The differences in the profile were greatest between co-occurring species and between closely related sibling species. Lastly, there were significant differences in the chemical profile between two populations of a particular species. These findings provide key insights into the role chemical signals play in the attainment and maintenance of reproductive barriers between closely related co-occurring species.

Text-based person search via cross-modal alignment learning

Text-based person search aims to use text descriptions to search for corresponding person images. However, due to the obvious pattern differences in image and text modalities, it is still a challenging problem to align the two modalities. Most existing approaches only consider semantic alignment within a global context or partial parts, lacking consideration of how to match image and text in terms of differences in model information. Therefore, in this paper, we propose an efficient Modality-Aligned Person Search network (MAPS) to address this problem. First, we suppress image-specific information by image feature style normalization to achieve modality knowledge alignment and reduce information differences between text and image. Second, we design a multi-granularity modal feature fusion and optimization method to enrich the modal features. To address the problem of useless and redundant information in the multi-granularity fused features, we propose a Multi-granularity Feature Self-optimization Module (MFSM) to adaptively adjust the corresponding contributions of different granularities in the fused features of the two modalities. Finally, to address the problem of information inconsistency in the training and inference stages, we propose a Cross-instance Feature Alignment (CFA) to help the network enhance category-level generalization ability and improve retrieval performance. Extensive experiments demonstrate that our MAPS achieves state-of-the-art performance on all text-based person search datasets, and significantly outperforms other existing methods.

Towards Compact 3D Representations via Point Feature Enhancement Masked Autoencoders

Learning 3D representation plays a critical role in masked autoencoder (MAE) based pre-training methods for point cloud, including single-modal and cross-modal based MAE. Specifically, although cross-modal MAE methods learn strong 3D representations via the auxiliary of other modal knowledge, they often suffer from heavy computational burdens and heavily rely on massive cross-modal data pairs that are often unavailable, which hinders their applications in practice. Instead, single-modal methods with solely point clouds as input are preferred in real applications due to their simplicity and efficiency. However, such methods easily suffer from limited 3D representations with global random mask input. To learn compact 3D representations, we propose a simple yet effective Point Feature Enhancement Masked Autoencoders (Point-FEMAE), which mainly consists of a global branch and a local branch to capture latent semantic features. Specifically, to learn more compact features, a share-parameter Transformer encoder is introduced to extract point features from the global and local unmasked patches obtained by global random and local block mask strategies, followed by a specific decoder to reconstruct. Meanwhile, to further enhance features in the local branch, we propose a Local Enhancement Module with local patch convolution to perceive fine-grained local context at larger scales. Our method significantly improves the pre-training efficiency compared to cross-modal alternatives, and extensive downstream experiments underscore the state-of-the-art effectiveness, particularly outperforming our baseline (Point-MAE) by 5.16%, 5.00%, and 5.04% in three variants of ScanObjectNN, respectively. Code is available at https://github.com/zyh16143998882/AAAI24-PointFEMAE.

The knowledge, attitudes and perceptions of physiotherapists and chiropractors in South Africa.

Effective healthcare delivery occurs when health professionals collaborate and provide holistic, patient-centred care. Physiotherapists and chiropractors treat a common range of patients with an overlap in their scope of practice and modalities because of typical healthcare roles that could lead to 'perceived' animosity. To assess the knowledge, attitudes, and perceptions of qualified chiropractors and physiotherapists regarding each other's practice. A cross-sectional survey using an online questionnaire and analysed descriptively. Participants were chiropractors (n = 116) and physiotherapists (n = 190). Chiropractors achieved a mean knowledge score of 75.7%, with physiotherapists at 59.7% on the assessments of each other's patients; an average score of 85.3% and 72.0% respectively, on knowledge of treatment modalities; knowledge score of 82.4% and 77.3% respectively, on the conditions treated by the other professional. A total of 82.8% (n = 96) of chiropractors and 70.0% (n = 133) of physiotherapists indicated the other professionals' competence in treating neuromusculoskeletal conditions. Inter-professional referrals occurred between 81.9% of chiropractors (n = 95) and 55.3% of physiotherapists (n = 105). Chiropractors (69.0%, n = 80) and physiotherapists (55.3%, n = 105) wanted to collaborate to manage patients. In the surveyed population in South Africa, chiropractors and physiotherapists had good knowledge, positive attitudes and perceptions of each other's practices, especially in the private sector. Inter-professional collaboration between chiropractors and physiotherapists should be encouraged so that healthcare delivery can be holistic and patient-centred for better clinical outcomes.

Cross‐modal knowledge learning with scene text for fine‐grained image classification

AbstractScene text in natural images carries additional semantic information to aid in image classification. Existing methods lack full consideration of the deep understanding of the text and the visual text relationship, which results in the difficult to judge the semantic accuracy and the relevance of the visual text. This paper proposes image classification based on Cross modal Knowledge Learning of Scene Text (CKLST) method. CKLST consists of three stages: cross‐modal scene text recognition, text semantic enhancement, and visual‐text feature alignment. In the first stage, multi‐attention is used to extract features layer by layer, and a self‐mask‐based iterative correction strategy is utilized to improve the scene text recognition accuracy. In the second stage, knowledge features are extracted using external knowledge and are fused with text features to enhance text semantic information. In the third stage, CKLST realizes visual‐text feature alignment across attention mechanisms with a similarity matrix, thus the correlation between images and text can be captured to improve the accuracy of the image classification tasks. On Con‐Text dataset, Crowd Activity dataset, Drink Bottle dataset, and Synth Text dataset, CKLST can perform significantly better than other baselines on fine‐grained image classification, with improvements of 3.54%, 5.37%, 3.28%, and 2.81% over the best baseline in mAP, respectively.

Husserl on knowing essences: Transworld identity and epistemic progression

AbstractHusserl's proposed method for knowing the essences of universals, which he calls “free variation,” has been widely criticized for involving viciously circular reasoning. In this paper, I review existing attempts to resolve this problem, and I argue that they all fail. I then show that extant accounts are all guilty of a common mistake: they assume that circularity is inevitable as long as the exercise of free variation presupposes the ability to identify the universal whose essence is in question, that is, the ability to recognize entities as instances of it. I reject this assumption: I argue on both Husserlian and independent philosophical grounds that knowledge of a universal's essence is not required for identifying it, but only for re‐identifying it at every possible world in which it is instantiated. I then defend a reading on which free variation's purpose is to move its practitioner from non‐essentialistic knowledge of a universal's identity (its actual instantiation‐pattern) to essentialistic knowledge of its transworld identity (its instantiation‐pattern in every possible world in which it is present). And I show that such a transformation is a non‐circular progression from non‐modal to modal knowledge.

Knowledge of community pharmacists on infantile haemangioma: Gaps and lack confidence to deliver propranolol still need to be filled.

To evaluate the knowledge, practices and self-confidence of community pharmacists, pharmacy technicians and pharmacy students about infantile haemangioma (IH) and propranolol treatment. A national survey was conducted in France from May 2022 to October 2022. A 42-item online questionnaire was used to assess pharmacists' knowledge of the epidemiology, clinical features and management of IH and propranolol treatment. The survey included 255 participants. The mean age was 34.9 years (±9.0); 225 (88%) were women. In all, 193 (76%) practised in urban pharmacies. Altogether, 83 participants (33%) had delivered oral propranolol solution for IH in the last 6 months. Participants' median score for self-confidence regarding propranolol dispensing was five (interquartile range, 2.5-6) on a scale of 1 to 10. Overall, 96 (38%) had more than 50% correct answers on the questionnaire. Multinomial regression models showed high scores on the questionnaire associated with high self-confidence when delivering oral propranolol solution, low number of years since graduation and having already delivered propranolol treatment. This study highlights a lack of knowledge of IH and modalities of propranolol treatment by community pharmacists and slight self-confidence when delivering propranolol. Greater cooperation between healthcare professionals could improve the proper use of medicine.

Attention-based 3D convolutional recurrent neural network model for multimodal emotion recognition.

Multimodal emotion recognition has become a hot topic in human-computer interaction and intelligent healthcare fields. However, combining information from different human different modalities for emotion computation is still challenging. In this paper, we propose a three-dimensional convolutional recurrent neural network model (referred to as 3FACRNN network) based on multimodal fusion and attention mechanism. The 3FACRNN network model consists of a visual network and an EEG network. The visual network is composed of a cascaded convolutional neural network-time convolutional network (CNN-TCN). In the EEG network, the 3D feature building module was added to integrate band information, spatial information and temporal information of the EEG signal, and the band attention and self-attention modules were added to the convolutional recurrent neural network (CRNN). The former explores the effect of different frequency bands on network recognition performance, while the latter is to obtain the intrinsic similarity of different EEG samples. To investigate the effect of different frequency bands on the experiment, we obtained the average attention mask for all subjects in different frequency bands. The distribution of the attention masks across the different frequency bands suggests that signals more relevant to human emotions may be active in the high frequency bands γ (31-50 Hz). Finally, we try to use the multi-task loss function Lc to force the approximation of the intermediate feature vectors of the visual and EEG modalities, with the aim of using the knowledge of the visual modalities to improve the performance of the EEG network model. The mean recognition accuracy and standard deviation of the proposed method on the two multimodal sentiment datasets DEAP and MAHNOB-HCI (arousal, valence) were 96.75 ± 1.75, 96.86 ± 1.33; 97.55 ± 1.51, 98.37 ± 1.07, better than those of the state-of-the-art multimodal recognition approaches. The experimental results show that starting from the multimodal information, the facial video frames and electroencephalogram (EEG) signals of the subjects are used as inputs to the emotion recognition network, which can enhance the stability of the emotion network and improve the recognition accuracy of the emotion network. In addition, in future work, we will try to utilize sparse matrix methods and deep convolutional networks to improve the performance of multimodal emotion networks.

Understanding perspectives on neural tube defect management: insights from Jordanian parents

Introduction Neural tube defects (NTDs) represent a spectrum of heterogeneous birth anomalies characterized by the incomplete closure of the neural tube. In Jordan, NTDs are estimated to occur in approximately one out of every 1000 live births. Timely identification of NTDs during the 18–22 weeks of gestation period offers parents various management options, including intrauterine NTD repair and termination of pregnancy (TOP). This study aims to assess and compare parental knowledge and perceptions of these management modalities between parents of affected children and those with healthy offspring. Materials and methods This retrospective case-control study was conducted at Jordan University Hospital (JUH) using telephone-administered questionnaires. Categorical variables were summarized using counts and percentages, while continuous variables were analyzed using mean and standard deviation. The association between exposure variables and outcomes was explored using binary logistic regression. Data analysis was performed using SPSS for Windows version 26 (SPSS Inc., Chicago, IL). Results The study sample comprised 143 participants, with 49.7% being parents of children with NTDs. The majority of NTD cases were associated with unplanned pregnancies, lack of folic acid supplementation, and postnatal diagnosis. Concerning parental knowledge of TOP in Jordan, 86% believed it to be legally permissible in certain situations. However, there was no statistically significant difference between cases and controls regarding attitudes toward TOP. While the majority of parents with NTD-affected children (88.7%) expressed a willingness to consider intrauterine surgery, this percentage decreased significantly (to 77.6%) after receiving detailed information about the procedure’s risks and benefits (p = .013). Conclusions This study represents the first case-control investigational study in Jordan focusing on parental perspectives regarding TOP versus intrauterine repair of myelomeningocele following a diagnosis of an NTD-affected fetus. Based on our findings, we urge the implementation of a national and international surveillance program for NTDs, assessing the disease burden, facilitating resource allocation toward prevention strategies, and promoting early diagnosis initiatives either by using newly suggested diagnostic biomarkers or early Antenatal ultrasonography.

Atypical antipsychotic-induced neuroleptic malignant syndrome

The rare neuroleptic malignant syndrome (NMS) is associated with early mortality of >30%. Better supportive care has reduced mortality to <10%. Atypical antipsychotics such as clozapine and risperidone are associated with an atypical but potentially lethal, NMS. A 22-year-old male developed NMS after being treated with clozapine and risperidone for aggression associated with mental retardation for about 8 years. For the past 1 year, the dose of clozapine was 200 mg/day and the dose of risperidone was 8 mg/day. The patient presented with symptoms of hyperthermia, perspiration, neck rigidity, and tonic posturing of all four limbs with tongue bite and frothing from the mouth. On examination, the patient’s Glasgow coma Scale was E3V1M4. He was intubated and given antibiotics, antiepileptic, tablet bromocriptine 2.5 mg (1-1-1-1), and other supportive treatment. The patient showed improvement in 10 days. Knowledge of diagnostic features, risk factors, treatment modalities, and differential diagnosis of NMS is essential. Early diagnosis and prompt treatment can reduce mortality and morbidity significantly.

Three-dimensional echocardiography: technical aspects and imaging modalities

Real-time three-dimensional echocardiography (RT3DE) is increasingly available in the veterinary field due to continuous reduction in costs and improvement of equipment. Much like its motion-mode and bi-dimensional counterparts, acquisition and analysis of RT3DE images and datasets is greatly improved by a thorough understanding of the technological aspects, basic physic principles, and knowledge of available modalities with their advantages and drawbacks. In this review, the authors aim to describe how the currently available RT3DE technology has evolved, explain technical aspects of the equipment, and illustrate the most commonly available modalities for image acquisition and visualization.