Learning Points Research Articles

A General Deep Learning Point–Surface Fusion Framework for RGB Image Super-Resolution

Hyperspectral images are usually acquired in a scanning-based way, which can cause inconvenience in some situations. In these cases, RGB image spectral super-resolution technology emerges as an alternative. However, current mainstream spectral super-resolution methods aim to generate continuous spectral information at a very narrow range, limited to the visible light range. Some researchers introduce hyperspectral images as auxiliary data. But it is usually required that the auxiliary hyperspectral images have the same spatial range as RGB images. To address this issue, a general point–surface data fusion method is designed to achieve the RGB image spectral super-resolution goal in this paper, named GRSS-Net. The proposed method utilizes hyperspectral point data as auxiliary data to provide spectral reference information. Thus, the spectral super-resolution can extend the spectral reconstruction range according to spectral data. The proposed method utilizes compressed sensing theory as a fundamental physical mechanism and then unfolds the traditional hyperspectral image reconstruction optimization problem into a deep network. Finally, a high-spatial-resolution hyperspectral image can be obtained. Thus, the proposed method combines the non-linear feature extraction ability of deep learning and the interpretability of traditional physical models simultaneously. A series of experiments demonstrates that the proposed method can effectively reconstruct spectral information in RGB images. Meanwhile, the proposed method provides a framework of spectral super-resolution for different applications.

Identifying slow learners in an e-learning environment using k-means clustering approach

Currently, the majority of e-learning lessons created and disseminated advocate a “one-size-fits-all” teaching philosophy. The e-learning environment, however, includes slow learners in a noticeable way, just like in traditional classroom settings. Learning analytics of educational data from a learning management system (LMS) have been considered by the researchers as a potential means of identifying slow e-learners and supporting, contesting, and altering present educational practices in e-learning. We used the students’ rates of learning and grade points along with the total learning time, which is calculated from the time series log data, to cluster the learners. The rate at which a student learns determines whether he or she is a slow learner, an average learner, or a gifted learner. For classifying learners, we followed a step-by-step procedure that included instructional design to create a dataset, learning analytics of the dataset, and a machine learning strategy to cluster e-learners. The system has been adequately integrated with the methods for measuring student learning. A strategy based on the revised Bloom’s Taxonomy is offered for the assessment of learners. The K-Means clustering approach is used to group learners who have similar performance without collecting a learner’s previous academic records or demographic information. In the experimental evaluation, 7.7% of e-learners are grouped as slow learners, while advanced learners make up 61.3 percent of the student body and average learners make up 31 percent. According to the study, there is a correlation between learning rate and academic success, with fast learners having a lower learning rate.

Transcutaneous ultrasound level check for selective dorsal rhizotomy.

Single-level selective dorsal rhizotomy (SDR) surgery requires an intra-operative level check to identifythe L1 vertebral level or the conus medullaris. Typically, this requires a pre-operative or intra-operative x-ray. We present our experience using initial transcutaneous ultrasound as an alternative to x-ray level check. A prospective SDR database was used to identify patients. The operation notes were reviewed to identify the level check method and any complications or wrong-level surgery. Data are reported for the first 160 SDR surgeries performed within our centre, mean age 6.47 years (range 2.5-19 years). The first 11 patients had combined x-ray and transcutaneous ultrasound for pre-incision level check. This allowed the neurosurgeon to assess the accuracy and feasibility of using transcutaneous ultrasound instead of x-ray. The subsequent 149 patients had ultrasound alone for transcutaneous pre-incision level check. An intra-operative ultrasound level check was performed for all patients following skin incision and dissection down to the spinal lamina. In this way, the conus level was confirmed before dural opening. For all patients at all ages, the combination of initial transcutaneous ultrasound followed by intra-operative ultrasound allowed accurate identification of the conus. There were no instances of wrong-level surgery. Learning points are presented within this paper. Combined use of transcutaneous ultrasound followed by intra-operative ultrasound can allow accurate identification of the conus, saving radiation exposure and potentially improving theatre efficiency. Appropriate training and experience are required for any neurosurgeon using these techniques.

Adaptive learning point cloud and image diversity feature fusion network for 3D object detection

3D object detection is a critical task in the fields of virtual reality and autonomous driving. Given that each sensor has its own strengths and limitations, multi-sensor-based 3D object detection has gained popularity. However, most existing methods extract high-level image semantic features and fuse them with point cloud features, focusing solely on consistent information from both sensors while ignoring their complementary information. In this paper, we present a novel two-stage multi-sensor deep neural network, called the adaptive learning point cloud and image diversity feature fusion network (APIDFF-Net), for 3D object detection. Our approach employs the fine-grained image information to complement the point cloud information by combining low-level image features with high-level point cloud features. Specifically, we design a shallow image feature extraction module to learn fine-grained information from images, instead of relying on deep layer features with coarse-grained information. Furthermore, we design a diversity feature fusion (DFF) module that transforms low-level image features into point-wise image features and explores their complementary features through an attention mechanism, ensuring an effective combination of fine-grained image features and point cloud features. Experiments on the KITTI benchmark show that the proposed method outperforms state-of-the-art methods.

Learning from new colorectal cancers: a qualitative synthesis of significant event reports.

Colorectal cancer is the second leading cause of cancer-related mortality in the UK and a significant contributor to morbidity and mortality worldwide. Early diagnosis provides opportunities for intervention and improved survival. Significant event analysis (SEA) is a well-established quality improvement method for learning from new cancer diagnoses. To provide additional insights into diagnostic processes for colorectal cancer and to identify areas for improvement in patient care pathways. Fifty-three general practices across Pennine Lancashire, England, submitted one or more SEA reports as part of an incentivised scheme. A standardised data collection form was used to collate learning points and recommendations for improvements. In total, 161 reports were analysed using an inductive framework analysis approach. There was an overarching theme of building vigilance and collaboration between and within general practices and secondary care. The following four main sub-themes were also identified: education; individualised and flexible care; ownership and continuity; and communication. These findings provide additional insights into colorectal cancer pathways from a primary care perspective. Practices should be supported in developing protocols for assessment and follow-up of patients with varying presentations. Screening and access to investigations are paramount for improving early diagnosis; however, a flexible diagnostic approach is required according to the individual circumstances of each patient.

No Simple Solutions to Complex Problems: Cognitive Science Principles Can Guide but Not Prescribe Educational Decisions.

Cognitive science of learning points to solutions for making use of existing study and instruction time more effectively and efficiently. However, solutions are not and cannot be one-size-fits-all. This paper outlines the danger of overreliance on specific strategies as one-size-fits-all recommendations and highlights instead the cognitive learning processes that facilitate meaningful and long-lasting learning. Three of the most commonly recommended strategies from cognitive science provide a starting point; understanding the underlying processes allows us to tailor these recommendations to implement at the right time, in the right way, for the right content, and for the right students. Recommendations regard teacher training, the funding and incentivizing of educational interventions, guidelines for the development of educational technologies, and policies that focus on using existing instructional time more wisely.

Penerapan Metode Yanbu’a dalam Pembelajaran Membaca Al-Qur’an di TPQ Baiturrahmah Nagari Padang Laweh Kec. Sungai Pua Kab. Agam

The background to this research is because there are several problems that occur at TPQ Baiturrahmah Nagari Padang Laweh. Among them, teachers have implemented or used the Yanbu'a method when learning to read the Qur'an, but there is still a lack of knowledge or understanding of the students regarding the Yanbu'a method in learning to read the Qur'an and the students' lack of focus in the ongoing learning which is proven by many students who fussing, walking around talking to his friend. This research aims to find out how to apply the Yanbu'a method in learning to read the Al-Qur'an and the obstacles in applying the Yanbu'a method. This research is a qualitative descriptive study. The author uses data collection techniques including observation, documentation and interviews. The key informant in this research was the TPQ teacher, the main informant was TPQ student Baiturrahmah. The results of this research prove that the TPQ Baiturrahmah teacher in applying the Yanbu'a method in learning to read the Qur'an is in accordance with the steps of the Yanbu'a method, namely first, explaining the main points of learning, giving correct examples, listening to students' reading patiently, giving warnings if there are any. wrong reading, if the child is fluent and correct, the teacher adds pages, if the child is not fluent and correct, don't add pages and it's better to repeat, learning time is 60-70 minutes. Secondly, the obstacle that educators find in implementing the Yanbu'a method is that students are often lazy and busy indoors.

Laryngectomy in Young Patients: A Case Series and Review of the Literature.

Background and objective The peak incidence of laryngeal cancer is seen in individuals aged over 65 years, with very few patientsunder 50 years developing advanced laryngeal cancer necessitating laryngectomy. Apart from often delayed diagnosis, this younger cohort faces a unique set of challenges related to fertility preservation, lower recruitment to clinical trials, and significant psychological impact. In light of this, this case series aimed to examine the various characteristics ofpatientsbelow the age of 50 years undergoing total laryngectomy. Methods We reviewed departmental records at the University College Hospital London, spanning a period of 10 years, to identify patients who underwent total laryngectomy under the age of 50 years. Results The group comprised a total of nine patients over the age of 10 years: five males and four females. Six (66.7%) patients were smokers, and two (22.2%) had human papillomavirus (HPV)-16-positive disease. These patients underwent a variety of operative techniques. The length of postoperative inpatient stay varied greatly, ranging from five to 44 days (mean: 23 days). Conclusion There appears to be a lower prevalence of classical risk factors in our younger cohort undergoing total laryngectomy, as well as a reduced incidence of HPV-16 and a higher proportion of females. We also bring to light the significant psychological impact that these younger patients face and highlight the key learning point that clinicians must be vigilant in investigating younger patients with suspicious symptoms, even in the absence of obvious risk factors. Although further research is needed,this series is unique in that currently there are no other papers outlining laryngectomies in apatient group aged below 50 years.

TOWARDS ACCURATE INSTANCE SEGMENTATION IN LARGE-SCALE LIDAR POINT CLOUDS

Abstract. Panoptic segmentation is the combination of semantic and instance segmentation: assign the points in a 3D point cloud to semantic categories and partition them into distinct object instances. It has many obvious applications for outdoor scene understanding, from city mapping to forest management. Existing methods struggle to segment nearby instances of the same semantic category, like adjacent pieces of street furniture or neighbouring trees, which limits their usability for inventory- or management-type applications that rely on object instances. This study explores the steps of the panoptic segmentation pipeline concerned with clustering points into object instances, with the goal to alleviate that bottleneck. We find that a carefully designed clustering strategy, which leverages multiple types of learned point embeddings, significantly improves instance segmentation. Experiments on the NPM3D urban mobile mapping dataset and the FOR-instance forest dataset demonstrate the effectiveness and versatility of the proposed strategy.

Peer-to-peer energy sharing model considering multi-objective optimal allocation of shared energy storage in a multi-microgrid system

A novel peer-to-peer (P2P) energy sharing model incorporating shared energy storage (SES) is proposed in order to effectively utilize renewable energy sources and facilitate flexible energy trading among microgrids. The model is divided into three main blocks. In the first block, a multi-objective optimal allocation scheme for SES is developed to maximize SES benefits, minimize carbon emissions, and enhance the sustainability of the energy sharing zone (ESZ). The second block introduces a new internal pricing mechanism based on the energy supply and demand ratio. Further, an innovative optimal scheduling strategy for the ESZ is presented in the third block, which includes demand response strategies for residential microgrids and control strategies for SES, with the goal of minimizing the energy cost of a multi-microgrid system (MMS). To solve the proposed complex and constrained model, a multi-objective slime mould algorithm based on generalized opposition-based learning and reference points is proposed. Simulation results demonstrate that the introduction of SES and energy sharing can reduce MMS's energy cost by 89.45 % and increase the sustainability index of ESZ by 9.08 %. The proposed P2P trading mechanism can strictly lock the internal trading prices within the grid prices, which fully mobilizes the microgrids to participate in P2P trading.

Learning point cloud context information based on 3D transformer for more accurate and efficient classification

AbstractThe point cloud semantic understanding task has made remarkable progress along with the development of 3D deep learning. However, aggregating spatial information to improve the local feature learning capability of the network remains a major challenge. Many methods have been used for improving local information learning, such as constructing a multi‐area structure for capturing different area information. However, it will lose some local information due to the independent learning point feature. To solve this problem, a new network is proposed that considers the importance of the differences between points in the neighbourhood. Capturing local feature information can be enhanced by highlighting the different feature importance of the point cloud in the neighbourhood. First, T‐Net is constructed to learn the point cloud transformation matrix for point cloud disorder. Second, transformer is used to improve the problem of local information loss due to the independence of each point in the neighbourhood. The experimental results show that 92.2% accuracy overall was achieved on the ModelNet40 dataset and 93.8% accuracy overall was achieved on the ModelNet10 dataset.

Emerging from critique towards liberation: A Framework in leadership education.

What does liberation look like in leadership learning and education? This article offers examples for (re)imagining leadership education in program design, coordination, and assessment by centering the leadership for liberation framework and other liberatory approaches. The authors offer examples of how these frameworks serve as an entry point for college student liberatory leadership learning.

Domain Adaptive Sampling for Cross-Domain Point Cloud Recognition

Point cloud recognition has recently gained increasing research interest due to the huge potential in real-world applications such as autonomous driving, robotics, etc. However, the point clouds of similar objects often exhibit notable geometric variations due to the difference in capturing devices or environmental changes. This leads to significant performance degradation when the learned point cloud recognition model is applied to a new scenario, which is also known as the domain adaptation issue. In this work, we propose a new unsupervised domain adaptation approach for point cloud recognition via domain adaptive sampling (DAS). In particular, we propose a two-level sampling strategy of point level and instance level to improve the cross-domain recognition ability of the model. First, we propose a domain adaptive point sampling (DAPS) strategy to enhance the domain-invariant representation of point clouds by progressively focusing on representative points in each point cloud based on geometric consistency. Then, we further propose an instance-level domain adaptive cloud sampling (DACS) strategy to learn target-specific information based on a self-paced learning paradigm, where we select a set of pseudo-labeled target point clouds to train our designed light-weighted adapters without modifying the learned domain-invariant representation. We validate our domain adaptive sampling approach on the benchmark datasets PointDA-10 and GraspNetPC-10, where our method achieves new state-of-the-art performance.

Recommendation to implementation of remote patient monitoring in rheumatology: lessons learned and barriers to take

Remote patient monitoring (RPM) leverages advanced technology to monitor and manage patients’ health remotely and continuously. In 2022 European Alliance of Associations for Rheumatology (EULAR) points-to-consider for remote care were...

Outcomes and GP perceptions of reviewing patients at risk of harm from opioids

Abstract Introduction The Clinical Commissioning Group (CCG) Medicines Optimisation workplan for 2021-22 had a focus on reducing harm from opioids.1,2 In particular, asking GP practices to identify and review two cohorts of patients: those on a combination of oxycodone and amitriptyline or similar high-risk combinations, and those over the age of 65 prescribed fentanyl transdermal patches. Practices were then expected to consider actions on how to reduce the risk for patients and also to provide feedback to the CCG on how the reviews had progressed. Aim We aimed to understand the outcomes, challenges and learning points experienced by practices in reviewing these patients. Methods Sixty practices were invited to participate. A feedback form was received from each practice that completed either or both reviews (covering anonymised outcomes of each review, challenges encountered, and practice-level learning points / changes made). Quantitative data were collated for each action, and feedback comments were categorised into broad groups. No patient details were collected. As this project was a service evaluation, ethical approval was not required. Results Forty-three practices completed the review of the combination of oxycodone and amitriptyline, though 5 of these had no patients to review. Overall, 185 patients were identified. Of these, 21 (11%) patients had the combination prescribing ceased, 22 (12%) were commenced on a reduction plan, and another 46 (25%) have been made aware of the risk and to report respiratory symptoms. Forty-three practices completed the fentanyl review, though one of these had no patients to review. Overall, 272 patients over 65 years were identified with 64 (24%) invited for a review with their GP or other allied health professional, 39 (14%) patients were put on a fentanyl patch dose reduction plan, and 14 (5%) had prescribing ceased. Practices reported various difficulties encountered with these reviews, including patient engagement in making changes to their pain medication which patients are reluctant to reduce; counselling and raising awareness of long-term effects of opioids to help gain patient understanding; how to manage complex patients who had been stable on their regimen over many years; and that it was time consuming to regularly check in with the patient and plan reduction regimens. Changes made by practices included recording risks discussed with patient in their medical record for easier future review; ensuring regular reviews, including by practice-based pharmacists, of patients prescribed opioids; trying to make time to see patients face to face; raising awareness of the risks of prescribing amitriptyline alongside oxycodone e.g. serotonin syndrome; and recognising that a collaborative process and patient engagement is necessary. Discussion/Conclusion The actions undertaken by practices link into national recommendations,1 with some GP feedback similar to the challenges and opportunities to improve patient safety on opioid use as reported elsewhere.3 Even with this small targeted patient cohort, this work demonstrates difficulties encountered by GPs in refraining from opioid prescribing. Other limitations of this small-scale review include lack of certainty whether reductions in opioid prescribing were always clinically appropriate, and reliance on practice feedback.

The interplay between self-regulated learning behavioral factors and students’ performance in English language learning through Moodle

The increasing popularity of online learning environments, such as Moodle LMS, has led to a growing interest in identifying factors contributing to student success in language learning. Self-regulated learning behaviors, such as goal setting, planning, and self-monitoring, have been identified as key predictors of academic achievement. However, limited research on how these behaviors relate to success in online language learning environments is limited. This study aimed to identify self-regulation learning factors and assess behaviors in English language teaching through Moodle LMS by analyzing trace data. The study analyzed trace data from 1523 English language learners in a Moodle course and identified several behavioral factors. The final course point in English language learning is significantly predicted, including the number of completed quizzes, middle course points, engagement with course materials, time spent on tasks, completion score quizzes, access time in total, and pacing. The study found that completing quizzes was the strongest predictor of the final course point, followed by time spent on task, access time in total, and middle course point. The findings suggest that educators can use the identified behavioral factors to promote self-regulated learning online and develop interventions to support students struggling with self-regulated learning. The studies include using trace data to analyze behavioral patterns and focusing on self-regulated learning factors in online language learning. The study provides important insights into self-regulated learning factors and behaviors in English language learning through Moodle LMS, which can inform the development of effective interventions to support students in online language learning environments.

Compliance with Glasgow Modified Alcohol Withdrawal Scale Guidelines within Emergency Medicine Utilising Electronic Prescribing Protocols

Abstract Introduction Glasgow Modified Alcohol Withdrawal Scale (GMAWS) allows evidence-based management of patients at risk of alcohol withdrawals within the hospital setting. Adherence to this guideline mitigates risk of patients developing Wernicke-Korsakoff syndrome, a neurological disorder resulting from a thiamine deficiency induced by chronic alcohol consumption1. As Scotland has particularly high rates of alcohol misuse, utilisation of this guideline avoids development of this serious complication with a large at-risk population. However, non-compliance and incorrect prescribing has been prevalent as this can be difficult and time-consuming to manually prescribe leading to errors. This can be avoided by utilising the novel electronic prescribing protocol on Hospital Electronic Prescribing and Medicine Administration (HEPMA) programme. Aim The aim of this quality improvement project was to increase compliance with the GMAWS guideline and ensure correct prescribing within Emergency Medicine. Methods Data collection was supported by the HEPMA team, with daily reports of all identified patients to allow evaluation of GMAWS prescribing as per protocols. Identified patients were confirmed to be on GMAWS and not prescribed thiamine/benzodiazepines for another indication with presence of GMAWS paper chart. Data was initially collected over a two-week period to establish a baseline, with further two-week periods of data collection after interventions. Intervention 1 was distribution of a hospital wide memo regarding GMAWS protocol. Intervention 2 was buzzword teaching on GMAWS with small groups of prescribers. This study did not require ethics approval due to it being a service evaluation / improvement project. Results The data collection following interventions demonstrated improved compliance with HEPMA protocols, evidenced by an increase from 33.3% at baseline to 82.1% compliance. The intervention of buzzword teaching was most impactful with 82.1% compliance recorded following this. An overall increase of 55.6% in the total prescribing of GMAWS signifies an increased awareness of the guidelines and increased confidence with prescribing. Discussion/Conclusion As novel electronic prescribing continues to evolve, pharmacy is a main contributor for advocating optimised prescribing and supporting best practice. Significant learning points included that many patients were on unnecessarily prolonged courses of parental thiamine, which has significant cost and nursing time implications. This is supported by the fact that oral thiamine costs 0.5% of the total cost of giving parental thiamine daily and reduced risks from unnecessary prolonged cannulation. We were able to incorporate this learning into the buzzword teaching to prompt review of parental thiamine prescribed without a clear duration. This project has prompted a review of our health board’s GMAWS guideline to streamline and update guidance in line with updated NICE guidelines. Feedback from the buzzword teaching sessions has been incorporated to improve readability and aid use, via the rationalisation of the GMAWS paperwork. The corresponding HEPMA protocols will then be updated to reflect these changes. Furthermore, this work has influenced other specialities to create and teach the use of HEPMA protocols to optimise treatment.

Reformation of Indonesian Insolvency Regulation on Execution of Cross-Border Assets Based on the European Union

The current possibility of hassle-free negotiations and transactions across countries allows international trade to be a significant source of profit for businesses in Indonesia. In the process, businesses are often exposed to risks of insolvency for being unable to repay debts. An insolvent debtor owning assets in other countries shifts the case to a cross-border insolvency case. As there is currently an absence of cross-border insolvency regulation in Indonesia, this paper therefore aims to reform Indonesian insolvency regulation through analyzing and applying learning points from the European Union’s success in managing cross-border insolvency. This paper identifies as legal normative research and is done through prescriptive analysis. The data used is secondary data obtained through a literature review, which includes primary and secondary sources. This paper uses the statute approach and will be analyzed qualitatively. Through the discussion, it is suggested that Indonesia adopt several relevant aspects of the EU regulation. To ease implementation, Indonesia should consider moving towards the reformation as a part of ASEAN. Alternatively, Indonesia can also independently engage in bilateral or multilateral treaties. To add on, Indonesia should advance from the idea of absolute territorialism or absolute universalism and instead seek to implement modified universalism in its cross-border insolvency regulation.
 

A Pre-Procession Module for Point-Based Deep Learning in Dense Point Clouds in the Ship Engineering Field

Recently, point cloud technology has been applied in the ship engineering field. However, the dense point cloud acquired by terrestrial laser scanning (TLS) technology in ship engineering applications brings an obstacle to some powerful and advanced point-based deep learning point cloud processing methods. This paper presents a deep learning pre-procession module to ensure the feasibility of processing dense point clouds on commonly available computer devices. The pre-procession module is designed according to the traditional point cloud processing methods and the PointNet++ paradigm, and is evaluated on two ship structure datasets and two popular point cloud datasets. Experimental results illustrate that (i) the proposed module improves the performance of point-based deep learning semantic segmentation networks, and (ii) the proposed module empowers the existing point-based deep learning networks with the capability to process dense input point clouds. The proposed module may provide a useful semantic segmentation tool for realistic dense point clouds in various industrial applications.

Multi-nodal basin drainage in lower limb melanoma

Lymph node status is an important factor that influences outcomes in melanoma. Whilst certain anatomical areas have multiple-nodal basin drainage, limb melanomas are thought to have more predictable lymphatic drainage patterns, with lower limb melanomas reliably draining to the corresponding ipsilateral inguinal lymph node basin with occasional popliteal drainage. Here we share our unique experience of a patient with a lower limb melanoma demonstrating sentinel lymph nodes, and subsequent metastatic spread, in both the ipsilateral and contralateral inguinal lymph node basins, highlighting an important learning point with respect to our clinical examination of melanoma patients.