-
Frontiers in Oncology 2024Our previous studies have demonstrated that Raman spectroscopy could be used for skin cancer detection with good sensitivity and specificity. The objective of this study...
BACKGROUND
Our previous studies have demonstrated that Raman spectroscopy could be used for skin cancer detection with good sensitivity and specificity. The objective of this study is to determine if skin cancer detection can be further improved by combining deep neural networks and Raman spectroscopy.
PATIENTS AND METHODS
Raman spectra of 731 skin lesions were included in this study, containing 340 cancerous and precancerous lesions (melanoma, basal cell carcinoma, squamous cell carcinoma and actinic keratosis) and 391 benign lesions (melanocytic nevus and seborrheic keratosis). One-dimensional convolutional neural networks (1D-CNN) were developed for Raman spectral classification. The stratified samples were divided randomly into training (70%), validation (10%) and test set (20%), and were repeated 56 times using parallel computing. Different data augmentation strategies were implemented for the training dataset, including added random noise, spectral shift, spectral combination and artificially synthesized Raman spectra using one-dimensional generative adversarial networks (1D-GAN). The area under the receiver operating characteristic curve (ROC AUC) was used as a measure of the diagnostic performance. Conventional machine learning approaches, including partial least squares for discriminant analysis (PLS-DA), principal component and linear discriminant analysis (PC-LDA), support vector machine (SVM), and logistic regression (LR) were evaluated for comparison with the same data splitting scheme as the 1D-CNN.
RESULTS
The ROC AUC of the test dataset based on the original training spectra were 0.886±0.022 (1D-CNN), 0.870±0.028 (PLS-DA), 0.875±0.033 (PC-LDA), 0.864±0.027 (SVM), and 0.525±0.045 (LR), which were improved to 0.909±0.021 (1D-CNN), 0.899±0.022 (PLS-DA), 0.895±0.022 (PC-LDA), 0.901±0.020 (SVM), and 0.897±0.021 (LR) respectively after augmentation of the training dataset (p<0.0001, Wilcoxon test). Paired analyses of 1D-CNN with conventional machine learning approaches showed that 1D-CNN had a 1-3% improvement (p<0.001, Wilcoxon test).
CONCLUSIONS
Data augmentation not only improved the performance of both deep neural networks and conventional machine learning techniques by 2-4%, but also improved the performance of the models on spectra with higher noise or spectral shifting. Convolutional neural networks slightly outperformed conventional machine learning approaches for skin cancer detection by Raman spectroscopy.
PubMed: 38962264
DOI: 10.3389/fonc.2024.1320220 -
Frontiers in Plant Science 2024Growing grass-legume mixtures for forage production improves both yield productivity and nutritional quality, while also benefiting the environment by promoting species...
INTRODUCTION
Growing grass-legume mixtures for forage production improves both yield productivity and nutritional quality, while also benefiting the environment by promoting species biodiversity and enhancing soil fertility (through nitrogen fixation). Consequently, assessing legume proportions in grass-legume mixed swards is essential for breeding and cultivation. This study introduces an approach for automated classification and mapping of species in mixed grass-clover swards using object-based image analysis (OBIA).
METHODS
The OBIA procedure was established for both RGB and ten band multispectral (MS) images capturedby an unmanned aerial vehicle (UAV). The workflow integrated structural (canopy heights) and spectral variables (bands, vegetation indices) along with a machine learning algorithm (Random Forest) to perform image segmentation and classification. Spatial k-fold cross-validation was employed to assess accuracy.
RESULTS AND DISCUSSION
Results demonstrated good performance, achieving an overall accuracy of approximately 70%, for both RGB and MS-based imagery, with grass and clover classes yielding similar F1 scores, exceeding 0.7 values. The effectiveness of the OBIA procedure and classification was examined by analyzing correlations between predicted clover fractions and dry matter yield (DMY) proportions. This quantification revealed a positive and strong relationship, with R2 values exceeding 0.8 for RGB and MS-based classification outcomes. This indicates the potential of estimating (relative) clover coverage, which could assist breeders but also farmers in a precision agriculture context.
PubMed: 38962243
DOI: 10.3389/fpls.2024.1414181 -
Frontiers in Psychology 2024The value of music lies in its ability to evoke emotions. People can gain emotional experiences in music and can also regulate their own emotions through music. Music...
BACKGROUND
The value of music lies in its ability to evoke emotions. People can gain emotional experiences in music and can also regulate their own emotions through music. Music has its own structural rules, and exploring the relationship between musical structure and emotions is an important approach to understanding the mechanism of music-induced emotions. Musical mode refers to the arrangement of intervals around the tonic, presenting different musical modes based on the central tone and the arrangement of intervals, including Chinese pentatonic modes and Western major and minor modes. Musical morphology indicates significant differences in the construction intensity of traditional Chinese pentatonic modes and major and minor modes, affecting their mode forms and thus determining their adaptability to external influences.
AIMS
Exploring the modalities of music and the effects of individual music training experiences on emotion induction; validating whether musical modes exhibit cross-cultural universality in the process of emotion induction.
METHOD
This study recruited 65 university students as participants (34 with music training experience, 31 without music training experience). Through a passive listening paradigm using the GEMS and combined with a biofeedback equipment, it explored the differences in behavioral and physiological indicators (skin conductance, temperature, heart rate) of emotional experiences (basic and aesthetic emotions) influenced by the modal forms of Chinese traditional pentatonic modes and Western major and minor modes.
RESULTS
Firstly, the arousal level of music emotion is a primary factor influencing individuals' aesthetic emotional experiences in music, which is related to the intensity of modal construction in music; Secondly, the emotional pleasure and skin temperature change induced by pentatonic music are greater than those induced by major and minor modes; Thirdly, the arousal level, electrodermal change, and heart rate variability of major and minor modes are greater than those of pentatonic music; Finally, music training experience enhances college students' familiarity and preference for pentatonic music, thereby strengthening the electrodermal physiological indicators of emotional experiences.
CONCLUSION
The different modal forms of music express different levels of emotional arousal, leading to differences in individuals' emotional dimensions and physiological indicators in music. Additionally, individuals' music training experiences and cultural backgrounds also influence their experience of music emotions.
PubMed: 38962236
DOI: 10.3389/fpsyg.2024.1414014 -
Frontiers in Psychology 2024Ethical voice is a valuable ethical behavior that enables organizations to promptly recognize and rectify unethical issues and practices, thus preventing severe dilemmas...
INTRODUCTION
Ethical voice is a valuable ethical behavior that enables organizations to promptly recognize and rectify unethical issues and practices, thus preventing severe dilemmas and crises. Despite its importance, the extant literature has yet to fully explore the impact of a leader's ethical voice on subordinate outcomes. This study bridges this gap by integrating social identity theory and social exchange theory to scrutinize the process by which a leader's ethical voice affects subordinate task performance.
METHODS
We employ a serial mediation model to explore the mechanisms by which a leader's ethical voice enhances subordinates' task performance. Our theoretical framework is empirically validated using a dataset that includes 449 subordinate-leader pairings from Chinese enterprises.
RESULTS
The survey results demonstrate that a leader's ethical voice has a significant positive impact on subordinate task performance. Subordinate identification with leader and leader-member exchange not only individually mediate the effects of a leader's ethical voice on subordinate task behavior but also jointly serve as a chain-mediated mechanism in the influence of a leader's ethical voice on subordinate task behavior.
DISCUSSION
These findings illuminate the substantial effects that ethical leadership behaviors exert on employee performance and offer fresh perspectives on the intricate dynamics that govern this influence.
PubMed: 38962223
DOI: 10.3389/fpsyg.2024.1340769 -
Frontiers in Psychology 2024Dance has been proposed to support superior intrinsic motivation over non-dance forms of therapeutic physical activity. However, this hypothesis has yet to be evaluated...
Partnered dance evokes greater intrinsic motivation than home exercise as therapeutic activity for chemotherapy-induced deficits: secondary results of a randomized, controlled clinical trial.
INTRODUCTION
Dance has been proposed to support superior intrinsic motivation over non-dance forms of therapeutic physical activity. However, this hypothesis has yet to be evaluated empirically, particularly among populations living with neuropathology such as survivors of cancer with neurologic complications from chemotherapy treatment. Questions about motivation are relevant to clinical outcomes because motivation mediates neuroplasticity. We conducted this secondary analysis of a randomized-controlled study to begin to investigate the relationships between personal motivation and neurophysiologic effects of dance-based intervention for healthy aging among populations with neurologic complications of cancer.
METHODS
We measured motivation using the Intrinsic Motivation Inventory, a validated patient-reported outcome from the psychological approach of Self Determination Theory. We assessed intrinsic motivation, extrinsic motivation, and satisfaction with intervention within a randomized controlled trial of dance versus exercise designed to alleviate symptoms of chemotherapy-induced impairment. Fifty-two survivors of breast cancer with chemotherapy-induced neuropathy diagnosis and associated sensorimotor functional deficits were randomized (1:1) to 8 weeks of partnered dance or home exercise, performed biweekly (NCT05114005; R21-AG068831).
RESULTS
While satisfaction did not differ between interventions, intrinsic motivation was higher among participants randomized to dance than those randomized to exercise ( < 0.0001 at all timepoints: 2 weeks, 4 weeks, 6 weeks, and 8 weeks of intervention), as was extrinsic motivation at 2 weeks ( = 0.04) and 8 weeks ( = 0.01).
DISCUSSION
These data provide evidence that social dance is more motivating than the type of home exercise generally recommended as therapeutic physical activity. The results inform directions for future study of the effect of dance-based therapeutics on embodied agency, neuroplastic changes, and clinically-relevant neuropathic improvement.
PubMed: 38962217
DOI: 10.3389/fpsyg.2024.1383143 -
Frontiers in Psychology 2024Misophonia is commonly associated with negative emotional or physiological responses to specific sounds. However, the consensus definition emphasizes that misophonia...
Misophonia is commonly associated with negative emotional or physiological responses to specific sounds. However, the consensus definition emphasizes that misophonia entails much more than that. Even in cases of subclinical misophonia, where individuals do not meet the disorder criteria, the experience can still be burdensome, despite not currently causing significant distress or impairment. The S-Five is a psychometric tool for comprehensive assessment of five aspects of misophonic experience: internalizing, externalizing, impact, threat, and outburst, and includes S-Five-T section to evaluate feelings evoked by triggering sounds and their intensity. We examined whether the five-factor structure developed in the UK could be replicated in a Polish sample, including individuals with and without self-identified misophonia. The Polish version of the S-Five was translated and tested on 288 Polish-speaking individuals. Comprehensive psychometric evaluation, including factor structure, measurement invariance, test-retest reliability, internal consistency, and concurrent validity evaluations, was conducted on the translated scale. Exploratory factor analysis suggested similar structure to the original English study, while bootstrap exploratory graph analysis showed the factor structure to be reproducible in other samples. The scale was found to be bias free with respect to gender, internally consistent and stable in time, and evidence of validity was provided using MisoQuest and Misophonia Questionnaire. These results offer support for the cross-cultural stability of the five factors and provide preliminary evidence for the suitability of the Polish version for clinical and research purposes. The study also investigated five facets of misophonia, triggering sounds, emotional responses, and their associations with symptoms of psychopathology across various cultures. It underscores the central role of anger, distress, and panic, while also highlighting the mixed role of irritation and disgust in misophonia across different cultural contexts. Mouth sounds evoked the most pronounced reactions compared to other repetitive sounds, although there were discernible cultural differences in the nature and intensity of reactions to various trigger sounds. These findings hold significant implications for future research and underscore the importance of considering cultural nuances in both research and the clinical management of misophonia.
PubMed: 38962216
DOI: 10.3389/fpsyg.2024.1372870 -
Data in Brief Jun 2024The current work presents the generation of a comprehensive spatial dataset of a lightweight beam element composed of four twisted plywood strips, achieved through the...
The current work presents the generation of a comprehensive spatial dataset of a lightweight beam element composed of four twisted plywood strips, achieved through the application of Structure-from-Motion (SfM) - Multi-view Stereo (MVS) photogrammetry techniques in controlled laboratory conditions. The data collection process was meticulously conducted to ensure accuracy and precision, employing scale bars of varying lengths. The captured images were then processed using photogrammetric software, leading to the creation of point clouds, meshes, and texture files. These data files represent the 3D model of the beam at different mesh sizes (raw, high-poly, medium-poly, and low-poly), adding a high level of detail to the 3D visualization. The dataset holds significant reuse potential and offers essential resources for further studies in numerical modeling, simulations of complex structures, and training machine learning algorithms. This data can also serve as validation sets for emerging photogrammetry methods and form-finding techniques, especially ones involving large deformations and geometric nonlinearities, particularly within the structural engineering field.
PubMed: 38962210
DOI: 10.1016/j.dib.2024.110254 -
Data in Brief Jun 2024Monitoring ocean surface temperature is critical to infer the variability of the upper layers of the ocean, from short temporal scales to climatic change scales....
Monitoring ocean surface temperature is critical to infer the variability of the upper layers of the ocean, from short temporal scales to climatic change scales. Analysis of the climatological trends and anomalies is fundamental to comprehend the long-term effects of climate change on marine ecosystems and coastal regions. The original data for the dataset presented was collected by the Portuguese Hydrographic Institute () using seven Ondograph and Meteo-oceanography buoys anchored offshore along the Portuguese coast to acquire ocean surface temperatures. The original raw data was pre-processed to provide averages over 3-hour periods and daily averages, and this data constitutes the provided dataset. The 3-hour temperature averages were obtained mainly between 2011 and 2015, and the daily temperature averages were obtained in intervals that vary with the considered buoy, having an average interval of 14 years per buoy. The data gathered provides a considerable temporal window, enabling the creation of data series and the implementation of data mining algorithms to develop decision support systems. Collecting data makes it possible to validate simulated results obtained using approximation models. This allows for more accurate temperature readings and facilitates testing and correcting created models.
PubMed: 38962202
DOI: 10.1016/j.dib.2024.110287 -
Data in Brief Jun 2024This data article introduces a comprehensive dataset of real-world truck parking locations across Europe. The dataset comprises = 19,713 designated parking sites...
This data article introduces a comprehensive dataset of real-world truck parking locations across Europe. The dataset comprises = 19,713 designated parking sites classified according to public accessibility and suitability for heavy-duty trucks (HDTs). More specifically, core information comprises the truck stop category, latitude and longitude information, area size, and country assignment. Furthermore, additional information such as truck traffic flow volumes, proximity to the highway network, and land use information provide supplemental data on ambient conditions and thus enhance the contextual relevance of those locations. The dataset was systematically generated using OpenStreetMap (OSM) data, focusing on parking areas, rest areas, and fueling stations as predominant public truck parking sites. These locations were evaluated and filtered for truck accessibility and suitability and then complemented and validated using commercial truck routing / geocoding software. Further refinement was achieved by Mean-Shift clustering. The further integration of supplementary datasets increased the information level, and all clustered locations were labeled into four archetypal categories. Finally, filtering retained only confidently classified publicly accessible and truck-certified parking and service facilities. This dataset assists in finding real-world stop options for HDTs during national or international operations and identifying suitable and most attractive sites for deploying alternative charging or refueling infrastructures along the European transport network. Accordingly, it can serve as a valuable resource for research in traffic science, future energy systems, and alternative truck powertrains. Its added value extends to diverse stakeholders like Charge Point Operators (CPOs), truck manufacturers, logistics companies, and public authorities.
PubMed: 38962201
DOI: 10.1016/j.dib.2024.110277 -
Data in Brief Jun 2024Hyperspectral imaging, combined with deep learning techniques, has been employed to classify maize. However, the implementation of these automated methods often requires...
Hyperspectral imaging, combined with deep learning techniques, has been employed to classify maize. However, the implementation of these automated methods often requires substantial processing and computing resources, presenting a significant challenge for deployment on embedded devices due to high GPU power consumption. Access to Ghanaian local maize data for such classification tasks is also extremely difficult in Ghana. To address these challenges, this research aims to create a simple dataset comprising three distinct types of local maize seeds in Ghana. The goal is to facilitate the development of an efficient maize classification tool that minimizes computational costs and reduces human involvement in the process of grading seeds for marketing and production. The dataset is presented in two parts: raw images, consisting of 4,846 images, are categorized into bad and good. Specifically, 2,211 images belong to the bad class, while 2,635 belong to the good class. Augmented images consist of 28,910 images, with 13,250 representing bad data and 15,660 representing good data. All images have been validated by experts from Heritage Seeds Ghana and are freely available for use within the research community.
PubMed: 38962186
DOI: 10.1016/j.dib.2024.110261