Explore the vast range of titles available.

The automated detection of deceptive language is a crucial challenge in computational linguistics. This study provides a rigorous comparative analysis of three tiers of machine learning models for detecting instructed deception: traditional machine learning (SVM), fine-tuned discriminative models (BERT), and in-context learning with generalist Large Language Models (LLMs). Using the “cross-cultural deception detection” dataset, our findings reveal a clear performance hierarchy. While SVM performance is inconsistent, fine-tuned BERT models achieve substantially superior accuracy. Notably, a multilingual BERT model improves cross-topic accuracy on Spanish text to 90.14%, a gain of over 22 percentage points from its monolingual counterpart (67.20%). In contrast, modern LLMs perform poorly in zero-shot settings and fail to surpass the SVM baseline even with few-shot prompting, underscoring the effectiveness of task-specific fine-tuning. By transparently addressing the limitations of the solicited, low-stakes deception dataset, we establish a robust methodological baseline that clarifies the strengths of different modeling paradigms and informs future research into more complex, real-world deception phenomena.

Background: Amyotrophic lateral sclerosis (ALS) is a devastating disease that targets motor neurons. Upper motor neurons degeneration is pathologically characterized by brain iron accumulation. Signal attenuation in the shape of a ribbon at the posterior border of the precentral gyrus can be observed on conventional magnetic resonance imaging (MRI) sequences including T2-weighted sequence. Methods: With the aim to know the qualities of this potential marker of ALS, we conducted a prospective study. Patients with definite ALS in the age range of 40–70 years and healthy controls underwent 3T brain MRI using a standardized sequence. A second MRI was performed 18 months later under the same conditions in the patients with ALS. Results: Most of the patients with ALS (91.66%) exhibited a “black ribbon” (BR) with an average area of 79.98 mm3. Signal attenuation discriminated ALS with a mean value of 63.97 arbitrary units (AU) on the left BR (95% CI: 60.67–67.27), a mean value of 59.15 AU (95% CI: 54.78–63.53) on the right BR, and a significant difference with control subjects presenting a mean value of 107.85 AU (p < 0.001). The optimal cut-off point for differentiating patients with ALS from controls (sensitivity, 0.92; specificity, 0.93) was 83 AU. Forced vital capacity and muscle strength in the contralateral upper extremity were significantly correlated with the ribbon intensity in ALS. Patients who underwent a second study exhibited significant changes in the BR related to the rapid evolution of the disease. Conclusions: This marker represents a valuable tool for the selection of candidates and their follow-up in clinical trials. L’hypo-intensité du cortex moteur est-elle une des caractéristiques de la sclérose latérale amyotrophique? Contexte: La sclérose latérale amyotrophique (SLA) est une maladie dévastatrice qui affecte les neurones moteurs. Sur le plan pathologique, la dégénérescence de ces neurones est caractérisée par une accumulation de fer dans le cerveau. L’affaiblissement du potentiel nerveux lié à cette maladie prend la forme d’un ruban situé à la limite postérieure du gyrus précentral. Un tel symptôme peut être observé au moyen de séquences tirées d’examens d’IRM conventionnelle, ce qui inclut la pondération T2. Méthodes: Nous avons mené une étude prospective avec l’objectif de mieux connaître les propriétés de ce marqueur potentiel de la SLA. Ainsi, des patients clairement atteints de SLA et âgés de 40 à 70 ans, de même que des témoins en bonne santé, ont subi une IRM 3T en vertu d’une séquence conventionnelle. Dix-huit mois plus tard, dans les mêmes conditions, une deuxième IRM a été effectuée dans le cas des patients atteints de SLA. Résultats: La plupart de ces patients, soit 91,66 % d’entre eux, ont donné à voir une forme de « ruban noir » dont la superficie moyenne atteignait 79,98 mm3. L’affaiblissement du potentiel nerveux a aussi permis de distinguer la SLA, la valeur moyenne, en unités arbitraires (UA), étant de 63,97 pour la section gauche du « ruban noir » (IC 95 % : 60,67 - 67,27) et de 59,15 (IC 95 % : 54,78 - 63,53) pour la section droite de ce même ruban. Une différence notable a été observée par rapport aux témoins, ces derniers donnant à voir une valeur moyenne de 107,85 (UA) (p < 0,001). Pour distinguer les patients atteints de SLA et les témoins, nous avons aussi déterminé que la limite optimale d’inclusion se situait à 83 (UA) (sensibilité : 0,92 ; spécificité : 0,93). Tant la capacité vitale forcée (CVF) que la force musculaire du membre supérieur controlatéral se sont révélées fortement corrélées à l’intensité du ruban associé à la SLA. Les patients ayant participé à une deuxième étude ont quant à eux présenté des modifications importantes en ce qui a trait à ce ruban en raison de l’évolution accélérée de leur maladie. Conclusions: Ce marqueur représente donc un outil précieux en vue de la sélection et du suivi de candidats devant participer à des essais cliniques.

Phishing emails pose a significant threat to cybersecurity worldwide. There are already tools that mitigate the impact of these emails by filtering them, but these tools are only as reliable as their ability to detect new formats and techniques for creating phishing emails. In this paper, we investigated how traditional models and transformer models work on the classification task of identifying if an email is phishing or not. We realized that transformer models, in particular distilBERT, BERT, and roBERTa, had a significantly higher performance compared to traditional models like Logistic Regression, Random Forest, Support Vector Machine, and Naive Bayes. The process consisted of using a large and robust dataset of emails and applying preprocessing and optimization techniques to maximize the best result possible. roBERTa showed an outstanding capacity to identify phishing emails by achieving a maximum accuracy of 0.9943. Even though they were still successful, traditional models performed marginally worse; SVM performed the best, with an accuracy of 0.9876. The results emphasize the value of sophisticated text-processing methods and the potential of transformer models to improve email security by thwarting phishing attempts.

Cacao is a tropical tree species, is highly sensitive to soil water limitations, and its production is strongly affected by global climatic fluctuations. The study of variability in cacao germplasms offers insights into identifying morphological traits associated with drought tolerance. Assessing the genetic parameters of these traits is essential to optimize selection strategies in breeding programs. This study estimated genetic parameters for eighteen cacao progenies and seven parental clones in the field and in greenhouse conditions. The family heritability (h2fam) and individual heritability (h2ind) of pod production during the dry season were calculated as h2fam = 0.30 and h2ind = 0.27. Regarding relative water content (RWC), the highest h2fam value was observed after 180 min of dehydration (0.26). High genetic values were observed for the parental clones CEPEC-2002 and SJ-02, as well as their progenies, in terms of production and RWC in the field. The survival and RWC of greenhouse seedlings were correlated with the evaluations carried out in adult plants in the field (a maximum r of 0.49*** between RWC in the field and survival in the greenhouse and a maximum r of 0.66*** between RWC in both conditions). These findings provide valuable insights to enhance breeding strategies to improve cacao drought tolerance.

Introduction: Mild behavioral impairment (MBI), and particularly its impulse dyscontrol domain, has been linked to brain alterations suggestive of Alzheimer’s disease (AD) in subjects without dementia. We aimed to analyze the association of impulse dyscontrol with the loss of integrity in white-matter brain tracts in a group of cognitively normal older adults. Materials and Methods: Using linear regression models, we analyzed the effect of impulse dyscontrol scores (using the MBI-Checklist, MBI-C) on diffusivity metrics (fractional anisotropy, FA; mean diffusivity) controlling for age and sex in five white-matter regions of interest: cingulum, fornix, hippocampus, superior fronto-occipital fasciculus, and uncinate fasciculus. A total of 48 cognitively normal older adults were included in the study. Results: The mean age of the subjects was 67.5 years, and 28 (58.3%) were female. The mean impulse dyscontrol score was 2.9 (SD: 4.8, Rng: 0 –22). The effect of the impulse dyscontrol score, controlling for age and sex, in the diffusivity measures of regions of interest was only significant in the fornix FA after multiple-comparison correction (weighted least squares model β = –3.65 x 10 –3, SE = 1.27 x 10 –3, corrected p = 0.03, R2 = 0.31). Discussion: To our knowledge, this is the first time that an MBI domain is linked to MRI diffusivity measures in a group composed exclusively of cognitively normal older adults. Our findings add to the growing understanding of MBI, and particularly the impulse dyscontrol domain, as a potential behavioral marker indicating a higher risk for developing neurocognitive disorders. Materials and Methods: A prospective cross-sectional observational study was conducted on patients who underwent a cognitive evaluation between May and October 2022 at the memory unit of the Favaloro Foundation (Buenos Aires, Argentina) who had undergone a neuropsychological evaluation during the current year and were under the care of our cognitive neurology service. During the medical interview, patients were asked to show the objects they carry in their pockets (wallet, candy wrapper, cellphone, thread, loose coins, tickets, receipts, etc.). Each item was counted independently. The total number obtained was compared. The total number obtained was compared with the FAQ and CDR values of each patient. A correlation between these data was sought, and it was determined whether this relationship (number of items carried in the pocket/scale value) has predictive value to determine if the patient meets dementia criteria according to functionality (FAQ: 6 or more, CDR: 1 or more). Results: The sample consisted of 26 male patients (n = 26) with an average age of 69 years and MMSE Mean: 27.73. Within this group, 80.7% had a diagnosis of dementia (of any etiology) and the rest were either healthy patients or patients with mild cognitive impairment. The non- parametric MANN WHITNEY test was performed using the R statistical program. A statistically significant difference was found when comparing the dementia group and objects in the pocket (wilcox_test p 0.00854). When comparing the group of patients with more than three objects and the presence of dementia, significant p values were also found (wilcox_test p 0.0363). Conclusions: The number of objects in the pocket of patients with cognitive impairment could be a valuable tool for predicting dementia. The presence of 3 or more objects in the pocket of a patient with cognitive impairment could correlate with a low score on the functionality scales (FAQ less than 6, absence of dementia stage).

Background: Virtual reality (VR) has been successfully used as part of treatment for patients with chronic low back pain and kinesiophobia. However, the cost of VR systems is still too high to be used in interventions in developing countries, and systems may produce negative effects on patients, such as cybersickness and reduced sense of presence.Objective: This study presents a novel mobile-based extended reality (XR) system that combines views of the real world with a simulated video trajectory, tailored for people with kinesiophobia.Methods: This work presents the concept of the system, which is the first system aimed at reducing kinesiophobia that uses extended reality, as well as its development and an initial evaluation by kinesiology students. Furthermore, it shows the enhanced XR version, which takes into account feedback from both subjects and the medical team.Results: An XR low-cost system was developed and found to provide an acceptable user experience, but high cybersickness values.Conclusions: This research proposes, implements and tests a novel, low-cost alternative for the treatment of kinesiophobia related to chronic pain, representing the first system for these purposes that uses extended reality instead of virtual reality.

Although the presence of anosognosia in amnestic mild cognitive impairment (aMCI) may be predictive of conversion to Alzheimer’s disease (AD), little is known about its neural correlates in AD and aMCI. Four different groups were compared using volumetric and diffusion magnetic resonance imaging metrics in regions of interest (hippocampus and cingulum cortex gray matter, cingulum bundle white matter): aMCI subjects with anosognosia ( n = 6), aMCI subjects without anosognosia ( n = 12), AD subjects with anosognosia ( n = 6), and AD subjects without anosognosia ( n = 9). aMCI subjects with anosognosia displayed a significantly lower gray matter density (GMD) in the bilateral hippocampus than aMCI subjects without anosognosia, which was accounted for by bilateral hippocampal differences. Furthermore, we identified that the mean hippocampal gray matter density of aMCI subjects with anosognosia was not statistically different than that of AD subjects. The groups of aMCI and AD subjects with anosognosia also displayed a lower GMD in the bilateral cingulum cortex compared to subjects without anosognosia, but these differences were not statistically significant. No statistically significant differences were found in the fractional anisotropy or mean diffusivity of the hippocampus or cingulum between subjects with and without anosognosia in aMCI or AD groups. While these findings are derived from a small population of subjects and are in need of replication, they suggest that anosognosia in aMCI might be a useful clinical marker to suspect brain changes associated with AD neuropathology.

The cerebellar cognitive affective syndrome (CCAS) has been consistently described in patients with acute/subacute cerebellar injuries. However, studies with chronic patients have had controversial findings that have not been explored with new cerebellar-target tests, such as the CCAS scale (CCAS-S). The objective of this research is to prove and contrast the usefulness of the CCAS-S and the Montreal Cognitive Assessment (MoCA) test to evaluate cognitive/affective impairments in patients with chronic acquired cerebellar lesions, and to map the cerebellar areas whose lesions correlated with dysfunctions in these tests. CCAS-S and MoCA were administrated to 22 patients with isolated chronic cerebellar strokes and a matched comparison group. The neural bases underpinning both tests were explored with multivariate lesion-symptom mapping (LSM) methods. MoCA and CCAS-S had an adequate test performance with efficient discrimination between patients and healthy volunteers. However, only impairments determined by the CCAS-S resulted in significant regional localization within the cerebellum. Specifically, patients with chronic cerebellar lesions in right-lateralized posterolateral regions manifested cognitive impairments inherent to CCAS. These findings concurred with the anterior-sensorimotor/posterior-cognitive dichotomy in the human cerebellum and revealed clinically intra- and cross-lobular significant regions (portions of right lobule VI, VII, Crus I-II) for verbal tasks that overlap with the “language” functional boundaries in the cerebellum. Our findings prove the usefulness of MoCA and CCAS-S to reveal cognitive impairments in patients with chronic acquired cerebellar lesions. This study extends the understanding of long-term CCAS and introduces multivariate LSM methods to identify clinically intra- and cross-lobular significant regions underpinning chronic CCAS.

PurposeWe believed that the hippocampal complex is a fractal, and we try to demonstrate it.MethodsWe selected 12 magnetic resonance (MR) studies from healthy brains. Five women and 7 men without neurological or psychiatric disease were analyzed. The age range was50–71 years old, with a mean 57.1 ± 7.6 years. Image series of MR were obtainedthrough 3D SPGR in scanner GE Singa 3.0 T (general Electric, USA), matrixacquisition 512 Å–512 Å–120, field size 240 mm, voxel size 0.47 mm Å–0.47 mm Å–1.2 mm, repetition time of 13 ms and eco time of 5.6 ms. The MRI were imported inDICOM format in the OsiriXsoftware where they were contoured. Post processing wasdone with ImageJ software; Box Counting method for fractal analysis and we also usedthe Box Counting method of the BoneJ plugin.ResultsAll of the hippocampus analyzed were a fractal. The fractal dimension analysisdistribution of the measurements had a central tendency.The mean of hippocampal fractal dimension was 1.32635, the range was from 1.3373to 1.5344.We found a short interval of variability in the hippocampal fractal number.ConclusionsThe hippocampal complex is a fractal. The fractal analysis must be an objectivemeasurement that can help us as a descriptive tool in hippocampal anatomy andmaybe in a close future in the diagnosis of anatomical alterations.

Split ring resonators (SRRs) have been used extensively in metamaterials, showing a strong localization and enhancement of fields, which significantly improves the sensitivity and resolution of the electromagnetic field sensors. We propose the development of an electric field sensor for 2.4 GHz industrial, scientific, and medical (ISM) band applications, by modifying the renowned SRR to contain three concentric pairs of rings. The reduced size makes the sensor affordable for experiments by inserting it in phantoms in order to measure the specific absorption rate (SAR). Power was transmitted from a patch antenna to SRR, with a varying set of distances 1λ, 2λ, 3λ, or 5λ. Experimental measurements of power were conducted with and without a cylindrical distilled-water phantom with agar (4.54%) and NaCl (0.95%). We then computed the electric and magnetic fields and the SAR using these experimental readings of power for different distances. Our sensor was able to measure power values from 20 nW to 0.3 µW with no phantom, and 1 nW to 10 nW with a phantom, in accordance with the values reported for radiofrequency (RF) dosimetry. The sensitivity as a function of the distance determined for the specific case of a phantom was 0.3 µW/cm.