To fully characterize the orientation dependence of magnetization transfer (MT) and inhomogeneous MT (ihMT) measures in the whole white matter (WM), for both single‐fiber and crossing‐fiber voxels.

The availability of powerful open-source large language models (LLMs) opens exciting use-cases, such as automated personal assistants that adapt to the user's unique data and demands. Two key desiderata for such assistants are personalization-in the sense that the assistant should reflect the user's own style-and privacy-in the sense that users may prefer to always store their personal data locally, on their own computing device. We present a new design for such an automated assistant, for the specific use case of personal assistant for email generation, which we call Panza. Specifically, Panza can be both trained and inferenced locally on commodity hardware, and is personalized to the user's writing style. Panza's personalization features are based on a new technique called data playback, which allows us to fine-tune an LLM to better reflect a user's writing style using limited data. We show that, by combining efficient fine-tuning and inference methods, Panza can be executed entirely locally using limited resources-specifically, it can be executed within the same resources as a free Google Colab instance. Finally, our key methodological contribution is a careful study of evaluation metrics, and of how different choices of system components (e.g. the use of Retrieval-Augmented Generation or different fine-tuning approaches) impact the system's performance.

The results of orthodontic therapy largely depend, among other factors, on the preparation of the tooth enamel itself and the choice of material used to bond orthodontic brackets. The aim of this in vitro study was to determine the shear bond strength (SBS) and adhesive remnant index (ARI) score of thermo-cured glass–ionomers on different pretreated enamel, in comparison with the commonly used composite cement. Three commercially available nano-ionomer or highly viscous glass–ionomer cements (EQUIA Forte® Fil, EQUIA Fil, Ketac Universal) and two types of compo-sites (Heliosit Orthodontic, ConTec Go!) were investigated in this study. The research involved two hundred human premolars. The teeth were cleaned and polished, then randomly divided into five groups according to the enamel preparation method and the type of material. The enamel was treated in three different ways: polyacrylic acid, phosphoric acid, 5% NaOCl + etching with phosphoric acid, and a control group without treatment. Glass–ionomer cement was thermo-cured with heat from a polymerization unit during setting. Statistical analysis was performed using a Chi-square test and one-way ANOVA for independent samples. Spearman’s Rho correlation coefficient was used to examine the relationship. Regardless of the material type, the results indicated that the weakest bond between the bracket and tooth enamel was found in samples without enamel pretreatment. The majority of the materials stayed on the brackets in samples without enamel preparation, according to ARI scores. The study’s findings demonstrated that the strength of the adhesion between the bracket and enamel is greatly influenced by enamel etching and glass–ionomer thermo-curing. Clinical investigations would be required to validate the outcomes.

The availability of powerful open-source large language models (LLMs) opens exciting use-cases, such as using personal data to fine-tune these models to imitate a user's unique writing style. Two key requirements for such assistants are personalization - in the sense that the assistant should recognizably reflect the user's own writing style - and privacy - users may justifiably be wary of uploading extremely personal data, such as their email archive, to a third-party service. In this paper, we present a new design and evaluation for such an automated assistant, for the specific use case of email generation, which we call Panza. Panza's personalization features are based on a combination of fine-tuning using a variant of the Reverse Instructions technique together with Retrieval-Augmented Generation (RAG). We demonstrate that this combination allows us to fine-tune an LLM to reflect a user's writing style using limited data, while executing on extremely limited resources, e.g. on a free Google Colab instance. Our key methodological contribution is the first detailed study of evaluation metrics for this personalized writing task, and of how different choices of system components--the use of RAG and of different fine-tuning approaches-impact the system's performance. Additionally, we demonstrate that very little data - under 100 email samples - are sufficient to create models that convincingly imitate humans. This finding showcases a previously-unknown attack vector in language models - that access to a small number of writing samples can allow a bad actor to cheaply create generative models that imitate a target's writing style. We are releasing the full Panza code as well as three new email datasets licensed for research use at https://github.com/IST-DASLab/PanzaMail.

New accurate, precise, and sensitive spectrophotometric method were developed for the assay of L-ascorbic acid in pharmaceutical preparations. The determination of L-ascorbic acid was based on its oxidation by potassium peroxydisulfate in the presence of Ag(I) as a catalyst. The molar absorptivity of the proposed method was found to be 8.61 · 103 L mol-1 cm-1 at 248 nm. Beer's law was obeyed in the concentration range of 0.46–20.0 μg mL–1. Other compounds commonly found in vitamin C and multivitamin products did not interfere with the determination of L-ascorbic acid. The proposed method was successfully applied for the determination of L-ascorbic acid in pharmaceutical formulations. The results obtained with the proposed method showed good agreement with those given by the titrimetric method using iodine.

Abstract Objective Real-life management of patients with hypertension and chronic kidney disease (CKD) among European Society of Hypertension Excellence Centres (ESH-ECs) is unclear : we aimed to investigate it. Methods A survey was conducted in 2023. The questionnaire contained 64 questions asking ESH-ECs representatives to estimate how patients with CKD are managed. Results Overall, 88 ESH-ECS representatives from 27 countries participated. According to the responders, renin-angiotensin system (RAS) blockers, calcium-channel blockers and thiazides were often added when these medications were lacking in CKD patients, but physicians were more prone to initiate RAS blockers (90% [interquartile range: 70–95%]) than MRA (20% [10–30%]), SGLT2i (30% [20–50%]) or (GLP1-RA (10% [5–15%]). Despite treatment optimisation, 30% of responders indicated that hypertension remained uncontrolled (30% (15–40%) vs 18% [10%–25%]) in CKD and CKD patients, respectively). Hyperkalemia was the most frequent barrier to initiate RAS blockers, and dosage reduction was considered in 45% of responders when kalaemia was 5.5–5.9 mmol/L. Conclusions RAS blockers are initiated in most ESH-ECS in CKD patients, but MRA and SGLT2i initiations are less frequent. Hyperkalemia was the main barrier for initiation or adequate dosing of RAS blockade, and RAS blockers’ dosage reduction was the usual management. PLAIN LANGUAGE SUMMARY What is the context? Hypertension is a strong independent risk factor for development of chronic kidney disease (CKD) and progression of CKD to ESKD. Improved adherence to the guidelines in the treatment of CKD is believed to provide further reduction of cardiorenal events. European Society of Hypertension Excellence Centres (ESH-ECs) have been developed in Europe to provide excellency regarding management of patients with hypertension and implement guidelines. Numerous deficits regarding general practitioner CKD screening, use of nephroprotective drugs and referral to nephrologists prior to referral to ESH-ECs have been reported. In contrast, real-life management of these patients among ESH-ECs is unknown. Before implementation of strategies to improve guideline adherence in Europe, we aimed to investigate how patients with CKD are managed among the ESH-ECs. What is the study about? In this study, a survey was conducted in 2023 by the ESH to assess management of CKD patients referred to ESH-ECs. The questionnaire contained 64 questions asking ESH-ECs representatives to estimate how patients with CKD are managed among their centres. What are the results? RAAS blockers are initiated in 90% of ESH-ECs in CKD patients, but the initiation of MRA and SGLT2i is less frequently done. Hyperkalemia is the main barrier for initiation or adequate dosing of RAAS blockade, and its most reported management was RAAS blockers dosage reduction. These findings will be crucial to implement strategies in order to improve management of patients with CKD and guideline adherence among ESH-ECs.

We present DeepRIoT, a continuous integration and continuous deployment (CI/CD) based architecture that accelerates the learning and deployment of a Robotic-IoT system trained from deep reinforcement learning (RL). We adopted a multi-stage approach that agilely trains a multi-objective RL controller in the simulator. We then collected traces from the real robot to optimize its plant model, and used transfer learning to adapt the controller to the updated model. We automated our framework through CI/CD pipelines, and finally, with low cost, succeeded in deploying our controller in a real F1tenth car that is able to reach the goal and avoid collision from a virtual car through mixed reality.

In everyday life, we make decisions in groups about a variety of issues. In group decision-making, group members discuss options, exchange preferences and opinions, and make a common decision. Decision support systems and group recommender systems facilitate this process by enabling preference elicitation, generating recommendations, and supporting the process. We are here interested in building a conversational system, namely, a chat app, enhanced with an AI agent supporting the group decision-making process. To design the system, rather than solely relying on our assumptions, we took one step back and conducted a comprehensive focus group study. This approach has allowed us to gain original insights into the specific needs and preferences of the future end-users, i.e., group members, ensuring that our system design aligns more closely with their requirements. The focus group study involved fourteen participants in three group compositions: friends, families, and couples. Our findings reveal that most of the group members define a good choice as one that maximizes overall satisfaction without leaving any member dissatisfied. Dealing with competing group members emerged as a primary concern, with study participants requesting specific help from the AI agent to address this challenge. Participants identified personality and group structure as crucial characteristics for the AI agent to properly operate, though some expressed privacy concerns. Lastly, participants expected an AI agent to provide private interactions with individual members, proactively guide discussions when necessary, continually analyze group interactions, and tailor support to those interactions.

Machine Learning-based heuristics have recently shown impressive performance in solving a variety of hard combinatorial optimization problems (COPs). However they generally rely on a separate neural model, specialized and trained for each single problem. Any variation of a problem requires adjustment of its model and re-training from scratch. In this paper, we propose GOAL (for Generalist combinatorial Optimization Agent Learning), a generalist model capable of efficiently solving multiple COPs and which can be fine-tuned to solve new COPs. GOAL consists of a single backbone plus light-weight problem-specific adapters for input and output processing. The backbone is based on a new form of mixed-attention blocks which allows to handle problems defined on graphs with arbitrary combinations of node, edge and instance-level features. Additionally, problems which involve heterogeneous types of nodes or edges are handled through a novel multi-type transformer architecture, where the attention blocks are duplicated to attend the meaningful combinations of types while relying on the same shared parameters. We train GOAL on a set of routing, scheduling and classic graph problems and show that it is only slightly inferior to the specialized baselines while being the first multi-task model that solves a wide range of COPs. Finally we showcase the strong transfer learning capacity of GOAL by fine-tuning it on several new problems. Our code is available at https://github.com/naver/goal-co/.

The purpose of the study was to evaluate and identifying the level of excess weight and obesity in older students between 15 and 18 years, as important benchmarks of the level of health in order to update the recommendations regarding the promotion of an active and healthy lifestyle. A cross-sectional study was conducted on a sample of 400 subjects, (186 boys and 214 girls), aged 15 to 18. Anthropometric data including: body height, body weight, Body Mass Index (BMI). Participants' BMI was estimated using the Percentile BMI calculator for children and teenagers aged 2 to 19. Study adolescents were defined as underweight, normal (healthy) weight, overweight, and obese according to the CDC child growth characteristics for age, sex, and BMI. 350 (85.5%) subjects were healthy weight; 26 respondents (6.5%) were overweight, 17 (4.25%), were obese, while 7 (1.75%) underweight. The analysis of the individual results of male and female subjects points to increased values of the body mass of males (18.81%), compared to female pupils (3%). Out of a total of 186 male students, 10.75% were in the overweight category, and 8.06% were categorized as obese, in constrast 2.80% of the girls were overweight and (1%<), in the obese category, which is an outstanding result, where obesity practically does not exist. According to the results of this study (for both sexes), in relation to gender, there were more malnourished girls (2.33%), compared to boys (1%<). Among high school students in Bosnia and Herzegovina, the number of children with overweight and obesity is relatively low compared to data from other countries. Based on the relevant results of this study, we consider it necessary to update strategies for promoting an active and healthy lifestyle regarding physical activity and eating habits for adolescents in relation to the specifics of the countries of residence and European trends. Keywords: BMI; students; overweight; obesity; weight status category; high school.

The purpose of this study was to evaluate the spatiotemporal immunoexpression pattern of microtubule-associated protein 1 light chain 3 beta (LC3B), glucose-regulated protein 78 (GRP78), heat shock protein 70 (HSP70), and lysosomal-associated membrane protein 2A (LAMP2A) in normal human fetal kidney development (CTRL) and kidneys affected with congenital anomalies of the kidney and urinary tract (CAKUT). Human fetal kidneys (control, horseshoe, dysplastic, duplex, and hypoplastic) from the 18th to the 38th developmental week underwent epifluorescence microscopy analysis after being stained with antibodies. Immunoreactivity was quantified in various kidney structures, and expression dynamics were examined using linear and nonlinear regression modeling. The punctate expression of LC3B was observed mainly in tubules and glomerular cells, with dysplastic kidneys displaying distinct staining patterns. In the control group’s glomeruli, LAMP2A showed a sporadic, punctate signal; in contrast to other phenotypes, duplex kidneys showed significantly stronger expression in convoluted tubules. GRP78 had a weaker expression in CAKUT kidneys, especially hypoplastic ones, while normal kidneys exhibited punctate staining of convoluted tubules and glomeruli. HSP70 staining varied among phenotypes, with dysplastic and hypoplastic kidneys exhibiting stronger staining compared to controls. Expression dynamics varied among observed autophagy markers and phenotypes, indicating their potential roles in normal and dysfunctional kidney development.

Machine Learning-based heuristics have recently shown impressive performance in solving a variety of hard combinatorial optimization problems (COPs). However they generally rely on a separate neural model, specialized and trained for each single problem. Any variation of a problem requires adjustment of its model and re-training from scratch. In this paper, we propose GOAL (for Generalist combinatorial Optimization Agent Learning), a generalist model capable of efficiently solving multiple COPs and which can be fine-tuned to solve new COPs. GOAL consists of a single backbone plus light-weight problem-specific adapters for input and output processing. The backbone is based on a new form of mixed-attention blocks which allows to handle problems defined on graphs with arbitrary combinations of node, edge and instance-level features. Additionally, problems which involve heterogeneous types of nodes or edges are handled through a novel multi-type transformer architecture, where the attention blocks are duplicated to attend the meaningful combinations of types while relying on the same shared parameters. We train GOAL on a set of routing, scheduling and classic graph problems and show that it is only slightly inferior to the specialized baselines while being the first multi-task model that solves a wide range of COPs. Finally we showcase the strong transfer learning capacity of GOAL by fine-tuning it on several new problems. Our code is available at https://github.com/naver/goal-co/.

Machine Learning-based heuristics have recently shown impressive performance in solving a variety of hard combinatorial optimization problems (COPs). However, they generally rely on a separate neural model, specialized and trained for each single problem. Any variation of a problem requires adjustment of its model and re-training from scratch. In this paper, we propose GOAL (for Generalist combinatorial Optimization Agent Learner), a generalist model capable of efficiently solving multiple COPs and which can be fine-tuned to solve new COPs. GOAL consists of a single backbone plus light-weight problem-specific adapters for input and output processing. The backbone is based on a new form of mixed-attention blocks which allows to handle problems defined on graphs with arbitrary combinations of node, edge and instance-level features. Additionally, problems which involve heterogeneous types of nodes or edges are handled through a novel multi-type transformer architecture, where the attention blocks are duplicated to attend the meaningful combinations of types while relying on the same shared parameters. We train GOAL on a set of routing, scheduling and classic graph problems and show that it is only slightly inferior to the specialized baselines while being the first multi-task model that solves a wide range of COPs. Finally we showcase the strong transfer learning capacity of GOAL by fine-tuning it on several new problems. Our code is available at https://github.com/naver/goal-co/.