Identical Training Research Articles

Enhancing self-directed learning with custom GPT AI facilitation among medical students: A randomized controlled trial

ABASTRACT Objective This study aims to assess the impact of LearnGuide, a specialized ChatGPT tool designed to support self-directed learning among medical students. Materials and methods In this 14-week randomized controlled trial (ClinicalTrials.gov NCT06276049), 103 medical students were assigned to either an intervention group, which received 12 weeks of problem-based training with LearnGuide support, or a control group, which received identical training without AI assistance. Primary and secondary outcomes, including Self-Directed Learning Scale scores at 6 and 12 weeks, Cornell Critical Thinking Test Level Z scores, and Global Flow Scores, were evaluated with a 14-week follow-up. Mann-Whitney U tests were used for statistical comparisons between the groups. Results At 6 weeks, the intervention group showed a marginally higher median Self-Directed Learning Scale score, which further improved by 12 weeks (4.15 [95% CI, 0.82 to 7.48]; p = 0.01) and was sustained at the 14-week follow-up. Additionally, this group demonstrated notable improvements in the Cornell Critical Thinking Test Score at 12 weeks (7.11 [95% CI, 4.50 to 9.72]; p < 0.001), which persisted into the 14-week follow-up. The group also experienced enhancements in the Global Flow Score from 6 weeks, maintaining superiority over the control group through 12 weeks. Conclusions LearnGuide significantly enhanced self-directed learning, critical thinking, and flow experiences in medical students, highlighting the crucial role of AI tools in advancing medical education.

Greater lactate accumulation does not alter peripheral concentrations of key appetite-regulating neuropeptides.

The potential mechanisms involved in lactate's role in exercise-induced appetite suppression requires further examination. We used sodium bicarbonate (NaHCO3) supplementation in a double-blind, placebo controlled, randomized crossover design to explore lactate's role on neuropeptide Y (NPY), agouti-related peptide (AgRP), and alpha-melanocyte stimulating hormone (α-MSH) concentrations. Twelve adults (7 males; 24.2±3.4 kg‧m-2; 42.18±8.56 mL‧kg-1‧min-1) completed two identical high-intensity interval training sessions following ingestion of NaHCO3 (BICARB) or sodium chloride (PLACEBO) pre-exercise. Blood lactate, acylated ghrelin, NPY, AgRP, α-MSH, and appetite perceptions were measured pre-exercise, 0-, 30-, 60-, and 90-min post-exercise. Free-living energy intake (electronic food diaries) was measured the day before, of, and after each experimental session. In BICARB, blood lactate was greater post-exercise (p<0.002, d>0.70) though acylated ghrelin was similar (p=0.075, =0.206) at all time-points post-exercise (p>0.034, d<0.22). NPY (p=0.006, >0.509) and AgRP (p<0.001, >0.488) had main effects of time increasing following exercise and returning to baseline, with no differences between sessions (NPY: p=0.0.192, =0.149; AgRP: p=0.422, =0.060). α-MSH had no main effect of time (p=0.573, =0.063) or session (p=0.269, =0.110). Appetite perceptions were similar during BICARB and PLACEBO (p=0.007, d=0.28) increasing in both sessions post-exercise (p<0.088, d>0.57). Energy intake had a main effect of day (p=0.025, =0.825), where the experimental session day was greater than the day before (p=0.010, d=0.59) with no other differences between days (p>0.260, d<0.38). The lower accumulation of lactate than our previous work did not generate exercise-induced appetite suppression as there were no differences in acylated ghrelin, appetite perceptions, or peripheral concentrations of neuropeptides.

Transcranial Direct Current Stimulation as Adjuvant to Gamified Rehabilitation for Upper Limb Function in Paediatric Brain Damage (CHILDBOOST Project): A Study Protocol for a Triple-Blind Randomised Controlled Trial

(1) Background and objectives: Paediatric brain injuries can lead to motor and cognitive deficits. Effective rehabilitation is critical for enhancing independence. While virtual reality (VR) and transcranial direct current stimulation (tDCS) have independently demonstrated beneficial effects on motor and cognitive functions, their combined efficacy and its cognitive effects remain to be explored in this population. We aim to investigate the effects of integrating tDCS with VR training on upper limb (UL) functionality and cognitive outcomes through a triple-blind randomised trial. (2) Methods: Twenty-eight children with hemiparesis secondary to non-progressive brain damage will be randomly allocated into two groups: active anodal tDCS (2 mA) plus UL VR training, and sham tDCS with identical VR training. The tDCS will target M1 of the affected or most affected hemisphere for 20 min, simultaneous to VR training. The following four assessments will be carried out: pre-intervention, post-intervention, and three- and six-months follow-up. (3) Results: This study will explore motor and cognitive outcomes of a motor-based intervention in paediatric brain damage. We hypothesise that the experimental group will show significant improvements in UL function and cognition, enhancing their functional recovery. (4) Conclusions: We propose a multidisciplinary therapeutic approach combining neuromodulation and VR to potentiate functional recovery through enhancing motor and cognitive performance in paediatric brain damage.

The effects of six weeks of combined balance and plyometric training on postural control performance in elite badminton players: A pilot randomized, controlled study

Background/ObjectivesThe athletic performance in badminton players largely depends on the capability of dynamic postural control to quickly restore balance when performing high-paced movements (e.g., frequent single-leg jumps). Our aim was to examine the effects of a novel intervention that combines balance training on an unstable surface and plyometric training on the performance of restoring balance after jumping, as well as related postural control in elite badminton players. MethodsSixteen elite male badminton players were randomly allocated to either a combined balance and plyometric training group (CT, n = 8) or a plyometric training group (PT, n = 8). The CT group participated in a six-week training program, which included three training sessions per week. Each session comprised 40 min of plyometric exercises and 20 min of balance training. The PT group underwent plyometric training using the identical protocol as that of the CT group. All participants underwent identical technical training in badminton throughout the duration of the study. At baseline and immediately after the intervention, participants completed a single-leg jumping test. The capacity to restore balance was evaluated using the time to stabilization (TTS) after landing; and the related center of pressure (COP) fluctuations were also recorded. The effect of intervention was examined by two-way repeated-measures of ANOVA. ResultsThe primary two-way repeated-measures ANOVA models showed no significant interactions between group and time on either the time to stability in the dominant leg (D-TTS) or the time to stability in the non-dominant leg (N-TTS) (p > 0.70). Significant main effects of time, group, and their interactions on dominant legs of the anterior-posterior displacement difference (D-COPAP) (time: p = 0.001; group: p = 0.001; interaction: p = 0.014), non-dominant legs of the anterior-posterior displacement difference (N-COPAP) (time: p < 0.001; group: p = 0.003; interaction: p = 0.021) and non-dominant legs of the medial-lateral displacement difference (N-COPML) (time: p < 0.001; group: p < 0.001; interaction: p = 0.026), that is, compared to baseline of both groups and post PT, the COP metrics were significantly reduced after CT. Secondarily, within the CT and PT group, after the intervention, the N-TTS (CT: p = 0.001, post: 0.58 ± 0.87; PT: p = 0.03, post: 0.71 ± 0.11) was significantly decreased compared to baseline (CT pre: 0.76 ± 0.16; PT pre: 0.88 ± 0.13). ConclusionThis pilot study demonstrated that, compared to PT-only, the 6-week CT which combines balance training induced comparable improvements in the capacity to restoring balance after landing from a single-leg jump, and significantly improved the postural control performance as measured by COP metrics.

Learning to Generalize With Object-Centric Agents in the Open World Survival Game Crafter

Reinforcement learning agents must generalize beyond their training experience. Prior work has focused mostly on identical training and evaluation environments. Starting from the recently introduced Crafter benchmark, a 2D open world survival game, we introduce a new set of environments suitable for evaluating some agent's ability to generalize on previously unseen (numbers of) objects and to adapt quickly (meta-learning). In Crafter, the agents are evaluated by the number of unlocked achievements (such as collecting resources) when trained for 1M steps. We show that current agents struggle to generalize, and introduce novel object-centric agents that improve over strong baselines. We also provide critical insights of general interest for future work on Crafter through several experiments. We show that careful hyper-parameter tuning improves the PPO baseline agent by a large margin and that even feedforward agents can unlock almost all achievements by relying on the inventory display. We achieve new state-of-the-art performance on the original Crafter environment. Additionally, when trained beyond 1M steps, our tuned agents can unlock almost all achievements. We show that the recurrent PPO agents improve over feedforward ones, even with the inventory information removed. We introduce CrafterOOD, a set of 15 new environments that evaluate OOD generalization. On CrafterOOD, we show that the current agents fail to generalize, whereas our novel object-centric agents achieve state-of-the-art OOD generalization while also being interpretable. Our code is public. <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/astanic/crafter-ood</uri>

Medical students and residents of both sexes have similar learning curves for laparoscopy simulator training

BackgroundSimulation-based laparoscopy training is a standard part of modern surgical education. Studies focused on optimizing training for surgical residents have often used medical students as study participants. This has been criticized as medical students might have different prerequisites for surgical skills training due to their limited knowledge and exposure to real-life laparoscopic procedures. Simulation-based training has also been criticized for favouring male trainees due to its resemblance with video games and focus on competition. This study aimed to investigate if there is any difference in skills acquisition between novice surgical residents and medical students for proficiency-based laparoscopic training. Furthermore, to examine if there was a difference between male and female participants. MethodWe did a post-hoc study combining data from 59 participants from four previously published randomized trials. We compared residents with medical students and females with male participants across studies with identical training programs. ResultsThe median time to reach proficiency in the first training program comparison was 295 (IQR: 269-357) and 318 (241-445) minutes, respectively, for the residents and medical students (p=0.87). Male trainees spent 323 (IQR: 281-371) minutes to reach proficiency compared to 318 (IQR: 241-445)minutes for females (p=0.73). In the second training program comparison, the residents and medical students spent 169 (IQR: 123-229) and 165 (IQR: 128-214) minutes (p=0.67), respectively, reaching proficiency. The times for males and females were 166 (IQR: 61-151)minutes and 169 (61-117) minutes, respectively (p=0.74). ConclusionSimilar learning curves were observed for laparoscopic skills acquisition for residents and medical students. No difference was observed between male and female participants for proficiency-based laparoscopy simulator training.

Point of Care Ultrasounds Obtained by Novice Physician Assistant Residents (POCUS ON PAR).

The integration of Point of Care Ultrasound (POCUS) into the care of trauma patients, specifically the E-FAST, has improved the accuracy of initial diagnoses and improved time to surgical intervention in critically ill patients. Physician assistants (PAs) are critically important members of any military trauma resuscitation team and are often team leaders in a pre-hospital setting. They may receive training in ultrasound but there are little data to support their use or evaluate their effectiveness in using POCUS. We designed a study to evaluate the image quality of an E-FAST Exam performed by Emergency Medicine Physician Assistant (EMPA) Fellows and Emergency Medicine (EM) Interns following identical training. Our hypothesis is that image quality obtained by EMPAs will be non-inferior to those images obtained by EM Interns. This is a prospective single-blinded study comparing the image quality of E-FAST exams performed by first year EM interns and first year EMPA fellows. All participants completed standard POCUS training prior to enrollment in the study. A total of 8 EMPAs and 8 EM first year residents completed 10 recorded E-FAST exams to be used as study images. Participants also viewed a 15-question slide show containing images of positive (6) and negative (9) E-FAST exams and recorded their interpretations. Images were reviewed by expert reviewers who were blinded to which images were collected by which group. An image quality score was recorded for each view as well as an overall image quality score. Image quality was rated on a 1 to 5 image quality scale. For overall image quality, the mean score for EMPAs was 3.6 ± 0.5 and for EM residents was 3.2 ± 0.5 with statistical significance favoring better image quality from the EMPAs. The time to completion for the EFAST exam for EMPAs was 4.8 ± 1.3 minutes and for interns it was 3.4 ± 1.4 minutes (P value = 0.02). There was no difference in image interpretation quiz scores between the groups (mean score 92% among interns and 95% among PAs). POCUS is an imaging modality which is very portable and relatively inexpensive which makes it ideal for military medicine. PAs are essential members of military trauma teams, and often run an initial trauma resuscitation. Being able to correctly identify patients who have free fluid early in the course of treatment allows for more correct evacuation criteria to ensure the sickest patients get to care the fastest. Although there are limited data to support POCUS use by non-physicians, our data support a growing body of evidence that it is not the profession or baseline medical education that determines an individual's ability to use and incorporate ultrasound into bedside and clinical practice. Our study shows that with training and experience PAs or other members of the military health care team can use the EFAST to better care for trauma patients.

The Effects of Resistance Training Interventions on Weight, Body Mass Index, Body Fat Percentage, and Flexibility in College Students: A Comparison Between Sports and Non-Sports Students

Background. Resistance training is an effective form of exercise that promotes healthy body weight regulation and enhances flexibility. However, discrepancies exist regarding the varying impacts of resistance training based on individuals’ training status across different fitness components, such as body composition and flexibility. Objectives. The primary objective of this study is to assess whether there are significant differences in the outcomes of resistance training responses between trained and untrained groups concerning body composition and flexibility. Materials and methods. This experimental resistance training study involved 60 male students (age = 20.83 ± 0.85 years old) from the same university. The participants were divided into two groups based on their training status: a trained group (S; n = 30) and an untrained group (NS; n = 30). Both groups underwent an identical training regimen, consisting of three sessions per week over a total of four weeks or 12 training sessions. Results. Significant differences were observed in the paired sample T-test analysis between pre-test and post-test results in the untrained group for body weight (t (29) = 7.940, p &lt; 0.001, d = 1.45), body mass index (t (29) = 7.579, p &lt; 0.001, d = 1.38), body fat percentage (t (29) = 9.733, p &lt; 0.001, d = 1.77), and sit-and-reach flexibility (t (29) = -7.714, p &lt; 0.001, d = -1.40). Similarly, the trained group showed significant changes in body weight (t (29) = 2.644, p = 0.013, d = 0.483), body fat (t (29) = 2.561, p = 0.016, d = 0.351), and flexibility (t (29) = -2.543, p = 0.017, d = 0.351), while no substantial difference was found in body mass index (t (29) = 1.925, p = 0.064, d = -0.464). Conclusion. The study found that using resistance training protocols targeting weight reduction with 60-70% of 1 Repetition Maximum (RM) over a one-month intervention period, without stringent calorie restriction, significantly decreased weight and BMI while enhancing flexibility. Notably, regardless of training status disparities, the untrained group demonstrated more expressed improvements compared to the trained group, indicating the influence of training status on response variations to training stimuli.

A moderated-mediation model of training transfer from leadership intervention

PurposeGrounding on the goal-setting theory and flow theory, this study explored the mechanism underlying the association between transfer design (TD); identical elements and training transfer (TT). Specifically, the authors explored a moderated mediation process of trainer performance and motivation to improve work through learning (MTIWL) that has received less consideration in the TT literature.Design/methodology/approachData were collected using the retro-perspective survey method. The first survey was administered offline (t1: the day when leadership intervention was completed. Subsequently, trainees were requested to participate in an online survey (t2: 12–14 weeks later). In all, 355 executives participated.FindingsThe results of structural equation modeling (SEM) analyses suggested that trainees’ MTIWL mediational impact between leadership intervention triggers (transfer design and identical elements), and TT was supported. In addition, the indirect impact of these variables on TT was found to be significant when the trainer had high performance than when it was low. This confirmed the trainer’s performance as a potential moderator in the TT process.Practical implicationsThis study is limited to the exploration of leadership intervention variables on TT. The findings have implications for leadership professionals and scholars who use leadership intervention and motivation metrics to predict TT.Originality/valueThis study offers a moderated mediation mechanism for enhancing TT through leadership intervention triggers. The proposed conceptual model included MTIWL as mediator and trainer performance during leadership intervention as moderator.

Multi-Echo Complex Quantitative Susceptibility Mapping and Quantitative Blood Oxygen Level-Dependent Magnitude (mcQSM + qBOLD or mcQQ) for Oxygen Extraction Fraction (OEF) Mapping.

Oxygen extraction fraction (OEF), the fraction of oxygen that tissue extracts from blood, is an essential biomarker used to directly assess tissue viability and function in neurologic disorders. In ischemic stroke, for example, increased OEF can indicate the presence of penumbra-tissue with low perfusion yet intact cellular integrity-making it a primary therapeutic target. However, practical OEF mapping methods are not currently available in clinical settings, owing to the impractical data acquisitions in positron emission tomography (PET) and the limitations of existing MRI techniques. Recently, a novel MRI-based OEF mapping technique, termed QQ, was proposed. It shows high potential for clinical use by utilizing a routine sequence and removing the need for impractical multiple gas inhalations. However, QQ relies on the assumptions of Gaussian noise in susceptibility and multi-echo gradient echo (mGRE) magnitude signals for OEF estimation. This assumption is unreliable in low signal-to-noise ratio (SNR) regions like disease-related lesions, risking inaccurate OEF estimation and potentially impacting clinical decisions. Addressing this, our study presents a novel multi-echo complex QQ (mcQQ) that models realistic Gaussian noise in mGRE complex signals. We implemented mcQQ using a deep learning framework (mcQQ-NET) and compared it with the existing QQ-NET in simulations, ischemic stroke patients, and healthy subjects, using identical training and testing datasets and schemes. In simulations, mcQQ-NET provided more accurate OEF than QQ-NET. In the subacute stroke patients, mcQQ-NET showed a lower average OEF ratio in lesions relative to unaffected contralateral normal tissue than QQ-NET. In the healthy subjects, mcQQ-NET provided uniform OEF maps, similar to QQ-NET, but without unrealistically high OEF outliers in areas of low SNR, such as SNR ≤ 15 (dB). Therefore, mcQQ-NET improves OEF accuracy by more accurately reflecting realistic Gaussian noise in complex mGRE signals. Its enhanced sensitivity to OEF abnormalities, based on more realistic biophysics modeling, suggests that mcQQ-NET has potential for investigating tissue variability in neurologic disorders.

Using cultural historical activity theory to understand how post-graduate residents perform discharge planning at a medical center in Taiwan

BackgroundDespite the importance of discharge planning in physicians’ education, currently in most countries, no identical training is provided. Difficulties in promoting physician discharge planning education in Taiwan are still noted. This study aims to find the physicians’ role of discharge planning training in educating post graduate year residents (PGY) in Taiwan.Materials and methodsWe took advantage of government and hospital policies that promote the discharge planning program to teach and implement it, beginning with PGY residents by incorporating it into their training program. We recruited 30 PGY residents who were attending their three-month general internal medicine training from 2018 to 2019. They were interviewed at the end of the program using cultural-historical activity theory (CHAT). Qualitative research methods were used to further understand how discharge planning and care was implemented.ResultsTrainees initially believed that they did not have any role in discharge planning. Using the cycle of expansive learning, we found that the role of physicians in discharge planning was unclear. There were still some inconsistencies in the teaching and implementation of the discharge planning program for PGY residents that needed to be resolved, but this study also let participants learn through practice to improve their identification of discharge planning.ConclusionsThis study analyzed the impact of a discharge planning program for PGY physicians in Taiwan. It showed that the program affected physicians’ practice and medical education, although some contradictions remain.

Improved deep learning image classification algorithm based on Swin Transformer V2

While convolutional operation effectively extracts local features, their limited receptive fields make it challenging to capture global dependencies. Transformer, on the other hand, excels at global modeling and effectively captures global dependencies. However, the self-attention mechanism used in Transformers lacks a local mechanism for information exchange within specific regions. This article attempts to leverage the strengths of both Transformers and convolutional neural networks (CNNs) to enhance the Swin Transformer V2 model. By incorporating both convolutional operation and self-attention mechanism, the enhanced model combines the local information-capturing capability of CNNs and the long-range dependency-capturing ability of Transformers. The improved model enhances the extraction of local information through the introduction of the Swin Transformer Stem, inverted residual feed-forward network, and Dual-Branch Downsampling structure. Subsequently, it models global dependencies using the improved self-attention mechanism. Additionally, downsampling is applied to the attention mechanism’s Q and K to reduce computational and memory overhead. Under identical training conditions, the proposed method significantly improves classification accuracy on multiple image classification datasets, showcasing more robust generalization capabilities.

Harmonized standard operating procedures for administering the ALS functional rating scale-revised

The ALS Functional Rating Scale-Revised is the most commonly used primary outcome measure in current ALS clinical trials. While rigorous training and certification is generally recognized as critical to reliable performance, differences have existed between training in the two groups responsible for most training in ALS outcome measures. We present a harmonized standard operating procedure which is intended to further reduce response variability by the use of identical training in North America and Europe.

The Effect of Acute Caffeine Intake on Resistance Training Volume, Prooxidant-Antioxidant Balance and Muscle Damage Markers Following a Session of Full-Body Resistance Exercise in Resistance-Trained Men Habituated to Caffeine

No previous study has analyzed the impact of caffeine intake on prooxidant-antioxidant balance and muscle damage following resistance exercise. The aim of this study was to determine the effect of 3 mg/kg of caffeine on the number of repetitions and the prooxidant-antioxidant balance and muscle damage after a session of full-body resistance exercise. Ten resistance-trained men habituated to caffeine participated in a randomized, crossover and double-blind experiment. Each participant performed two identical resistance training sessions after the intake of 3 mg/kg of caffeine or a placebo. Blood was collected before and 60 min after substance intake, just after exercise, 60 minutes after exercise, and 24 hours after testing to evaluate the activity of antioxidant enzymes (superoxide dismutase, glutathione peroxidase, catalase), non-enzymatic antioxidants (reduced glutathione, uric acid) levels of oxidative stress markers (plasma malondialdehyde) and muscle damage markers (creatine kinase, lactate dehydrogenase). There were no significant differences between placebo and caffeine conditions in the total number of repetitions (180 ± 15 vs 185 ± 14 repetitions, respectively; p = 0.276; Effect size [ES] = 0.34), the total time under tension (757 ± 71 vs 766 ± 56 s, respectively; p = 0.709; ES = 0.14) or the rating of perceived exertion (13.8 ± 2.7 vs 14.7 ± 2.7 a.u., respectively; p = 0.212; ES = 0.32). Reduced glutathione concentration obtained 1 hour after exercise was higher with caffeine than with placebo (p = 0.047), without significant difference between conditions for any other prooxidant-oxidant or muscle damage marker at any time point (p &gt; 0.050 for all). The oral intake of 3 mg/kg of caffeine by resistance-trained men habituated to caffeine did not enhance the number of repetitions during a medium load full-body resistance training session to failure and had a minimal impact on the prooxidant-antioxidant balance and muscle damage. The study was registered prospectively at ClinicalTrials.gov with the following ID: NCT05230303.

The Effect of Acute Caffeine Intake on Resistance Training Volume, Prooxidant-Antioxidant Balance and Muscle Damage Markers Following a Session of Full-Body Resistance Exercise in Resistance-Trained Men Habituated to Caffeine.

No previous study has analyzed the impact of caffeine intake on prooxidant-antioxidant balance and muscle damage following resistance exercise. The aim of this study was to determine the effect of 3 mg/kg of caffeine on the number of repetitions and the prooxidant-antioxidant balance and muscle damage after a session of full-body resistance exercise. Ten resistance-trained men habituated to caffeine participated in a randomized, crossover and double-blind experiment. Each participant performed two identical resistance training sessions after the intake of 3 mg/kg of caffeine or a placebo. Blood was collected before and 60 min after substance intake, just after exercise, 60 minutes after exercise, and 24 hours after testing to evaluate the activity of antioxidant enzymes (superoxide dismutase, glutathione peroxidase, catalase), non-enzymatic antioxidants (reduced glutathione, uric acid) levels of oxidative stress markers (plasma malondialdehyde) and muscle damage markers (creatine kinase, lactate dehydrogenase). There were no significant differences between placebo and caffeine conditions in the total number of repetitions (180 ± 15 vs 185 ± 14 repetitions, respectively; p = 0.276; Effect size [ES] = 0.34), the total time under tension (757 ± 71 vs 766 ± 56 s, respectively; p = 0.709; ES = 0.14) or the rating of perceived exertion (13.8 ± 2.7 vs 14.7 ± 2.7 a.u., respectively; p = 0.212; ES = 0.32). Reduced glutathione concentration obtained 1 hour after exercise was higher with caffeine than with placebo (p = 0.047), without significant difference between conditions for any other prooxidant-oxidant or muscle damage marker at any time point (p > 0.050 for all). The oral intake of 3 mg/kg of caffeine by resistance-trained men habituated to caffeine did not enhance the number of repetitions during a medium load full-body resistance training session to failure and had a minimal impact on the prooxidant-antioxidant balance and muscle damage. The study was registered prospectively at ClinicalTrials.gov with the following ID: NCT05230303.

Identifying suicide documentation in clinical notes through zero-shot learning.

In deep learning, a major difficulty in identifying suicidality and its risk factors in clinical notes is the lack of training samples given the small number of true positive instances among the number of patients screened. This paper describes a novel methodology that identifies suicidality in clinical notes by addressing this data sparsity issue through zero-shot learning. Our general aim was to develop a tool that leveraged zero-shot learning to effectively identify suicidality documentation in all types of clinical notes. US Veterans Affairs clinical notes served as data. The training data set label was determined using diagnostic codes of suicide attempt and self-harm. We used a base string associated with the target label of suicidality to provide auxiliary information by narrowing the positive training cases to those containing the base string. We trained a deep neural network by mapping the training documents' contents to a semantic space. For comparison, we trained another deep neural network using the identical training data set labels, and bag-of-words features. The zero-shot learning model outperformed the baseline model in terms of area under the curve, sensitivity, specificity, and positive predictive value at multiple probability thresholds. In applying a 0.90 probability threshold, the methodology identified notes documenting suicidality but not associated with a relevant ICD-10-CM code, with 94% accuracy. This method can effectively identify suicidality without manual annotation.

Training on Sand or Parquet: Impact of Pre-Season Training on Jumping, Sprinting, and Change of Direction Performance in Professional Basketball Players

This study examined the effects of combined change of direction (CoD) and plyometric training on sand in basketball players’ jump, sprint, and CoD performances. In total, twenty-five male professional basketball players (age, 24.1 ± 4.2 years; height, 192.5 ± 8.0 cm; body mass, 92.6 ± 11.6 kg) participated in a pre-season intervention study over seven weeks, where two teams completed an identical speed training protocol, either on sand or on a hard surface, while a third team served as the control. All participants followed their regular pre-season training, while the intervention groups additionally performed the training protocol twice weekly. A repeated-measures ANOVA (rANOVA) showed significant interaction effects on the countermovement jump (F = 14.90, p &lt; 0.001), pivot–step–jump test (F = 8.09, p = 0.002), 10 m sprint time (ST) (F = 4.18, p = 0.050), and 20 m ST (F = 8.49, p = 0.002). Moreover, an rANOVA showed significant interaction effects for the CoD performance regarding total time (F = 5.70, p = 0.010) and average velocity prior to the CoD (F = 8.40, p = 0.002) and after the CoD (F = 3.89, p = 0.036). As such, the findings suggest that sand-based training elicits kinematic adaptations, increased muscle activation, and a shift towards concentric force development that all contribute to enhanced athletic capabilities.

Heterogeneous-Branch Collaborative Learning for Dialogue Generation

With the development of deep learning, advanced dialogue generation methods usually require a greater amount of computational resources. One promising approach to obtaining a high-performance and lightweight model is knowledge distillation, which relies heavily on the pre-trained powerful teacher. Collaborative learning, also known as online knowledge distillation, is an effective way to conduct one-stage group distillation in the absence of a well-trained large teacher model. However, previous work has a severe branch homogeneity problem due to the same training objective and the independent identical training sets. To alleviate this problem, we consider the dialogue attributes in the training of network branches. Each branch learns the attribute-related features based on the selected subset. Furthermore, we propose a dual group-based knowledge distillation method, consisting of positive distillation and negative distillation, to further diversify the features of different branches in a steadily and interpretable way. The proposed approach significantly improves branch heterogeneity and outperforms state-of-the-art collaborative learning methods on two widely used open-domain dialogue datasets.

Effect of the identical training process on the functional state of high-level sprint swimmers of both genders

Purpose: to study the effect of an identical training process on the functional state of high-level sprint swimmers of both genders.&#x0D; Material and methods. The heart rate variability indices, central hemodynamic parameters and PWC170/kg measurements were compared between 86 male and 42 female swimmers with sports qualifications from first-class athlete to Master of Sports International Class, and separately, – between 38 males and 9 females (Masters of Sports and Masters of Sports International Class), 28 males and 16 females (Candidates for Masters of Sports), and 22 males and 17 females (first-class athletes).&#x0D; The results. In the general group (86 males vs. 42 females) only the greater value of PWC170/kg in males by 12.64 % (p&lt;0.001) was revealed, without any differences in heart rate variability and central hemodynamics. Masters of Soprts and Masters of Sports International Class, as well as Candidates for Masters of Sports, didn't have gender differences in all indicators studied. In first-class athletes, there were no gender differences in heart rate variability and central hemodynamic parameters, except for PWC170/kg, which in males was greater by 11.27 % (p=0.036).&#x0D; Conclusions. A comparison of 86 male and 42 female swimmers with sports qualifications from first-class athlete to Master of Sports International Class, who differed by age and training experience, showed no significant differences in HRV and central hemodynamic parameters except for PWC170/kg, which was 12.64 % higher in males (p&lt;0.001). There were no significant differences between all studied indices in 38 male and 9 female swimmers with sports qualifications Master of Sports and Master of Sports International Class, with no differences in age or training experience. Comparison of 28 male and 16 female swimmers with sports qualification Candidate for Master of Sports, not differing by age and training experience showed no significant differences in HRV and central hemodynamic parameters, except for PWC170/kg, which was greater in males by 11.27 % (p=0.034). In 22 male and 17 female swimmers with sports qualification first-class athlete, not differing by age (males have greater training experience) there were no differences between all studied indices.

Environmental heat stress offsets adaptation associated with carbohydrate periodization in trained male triathletes.

Carbohydrate (CHO) intake periodization via the sleep low train low (SL-TL) diet-exercise model increases fat oxidation during exercise and may enhance endurance-training adaptation and performance. Conversely, training under environmental heat stress increases CHO oxidation, but the potential of combined SL-TL and heat stress to enhance metabolic and performance outcomes is unknown. Twenty-three endurance-trained males were randomly assigned to either control (n = 7, CON), SL-TL (n = 8, SLTemp ) or SL-TL + heat stress (n = 8, SLHeat ) groups and prescribed identical 2-week cycling training interventions. CON and SLTemp completed all sessions at 20°C, but SLHeat at 35°C. All groups consumed matched CHO intake (6 g·kg-1 ·day-1 ) but timed differently to promote low CHO availability overnight and during morning exercise in both SL groups. Submaximal substrate utilization was assessed (at 20°C), and 30-min performance tests (at 20 and 35°C) were performed Pre-, Post-, and 1-week post-intervention (Post+1). SLTemp improved fat oxidation rates at 60% MAP (~66% VO2peak ) at Post+1 compared with CON (p < 0.01). Compared with SLTemp , fat oxidation rates were significantly lower in SLHeat at Post (p = 0.02) and Post+1 (p < 0.05). Compared with CON, performance was improved at Post in SLTemp in temperate conditions. Performance was not different between any groups or time points in hot conditions. SL-TL enhanced metabolic adaptation and performance compared with CON and combined SL-TL and heat stress. Additional environmental heat stress may impair positive adaptations associated with SL-TL.