Hasil untuk "Human anatomy"

D. Winter

M. Raichle, J. Fiez, T. Videen et al.

S. Gregor, L. Kruse, S. Seidel

This essay derives a schema for specifying design principles for information technology-based artifacts in sociotechnical systems. Design principles are used to specify design knowledge in an accessible form, but there is wide variation and lack of precision across views regarding their formulation. This variation is a sign of important issues that should be addressed, including a lack of attention to human actors and levels of complexity as well as differing views on causality, on the nature of the mechanisms used to achieve goals, and on the need for justificatory knowledge. The new schema includes the well-recognized elements of design principles, including goals in a specific context and the mechanisms to achieve the goal. In addition, the schema allows: (1) consideration of the varying roles of the human actors involved and the utility of design principles, (2) attending to the complexity of IT-based artifacts through decomposition, (3) distinction of the types of causation (i.e., deterministic versus probabilistic), (4) a variety of mechanisms in achieving aims, and (5) the optional definition of justificatory knowledge underlying the design principles. We illustrate the utility of the proposed schema by applying it to examples of published research.

T. A. Nguyen, M. Hyun, Jo et al.

Ming Zhang

Jahira Banu, Sulochana Sakthivel

Background: In the domains of forensic science, anatomy, archaeology, and sports medicine, morphometric data of long bones are important. The morphometry of the tibia is essential for the successful outcome of total knee arthroplasty. The lower end of the tibia, which forms the ankle joint, is commonly impacted in accidents and sports injuries. Objective: The present study aims to reappraise the morphometry of the adult dry tibia in South India. Materials and Methods: A cross-sectional observational study was conducted on 60 bones of adult human dry tibia obtained from the bone collection of the department of anatomy. Various parameters of the upper end, shaft, and lower end of the tibia were measured using a digital Vernier caliper. Results: The tibial length varied from 32.7 to 42.7 cm, with a mean length of 36.79 ± 2.24. The Cnemic index, Foraminal index, Length–thickness index, and Cross-section index were 72.57 ± 11, 32.1 ± 3.3, 75.7 ± 11.1, and 22.95 ± 1.53, respectively. According to the foraminal index, 75% of the bones belonged to Type-1 and 25% to Type-2, where the nutrient foramen was in the proximal and middle third of the bone, respectively. Conclusions: The present study gives observational morphometric data of the tibia, which will be helpful in clinical as well as in medicolegal cases. These parameters will help in the selection of the correct prosthesis for a successful surgery, and we hope this morphometric analysis will have substantial significance across various disciplines.

Aitana Vázquez‐Fernández, Francisco F. Caballero, Humberto Yévenes‐Briones et al.

ABSTRACT Background Multimorbidity is the most common chronic condition experienced among older adults. It is unknown which amount and source of protein influences the development of frailty and mortality in patients with multimorbidity. We aimed to examine the association of plant and animal sources of protein intake with frailty and mortality among this type of patients. Methods This longitudinal study included 1868 participants aged ≥ 60 years from the Seniors‐ENRICA cohort in Spain with multimorbidity, defined as having 2 or more clinician‐diagnosed chronic diseases. Habitual diet was assessed at baseline (2008–2010) with a validated computerized diet history. Participants underwent repeated physical examinations (in 2013, 2015 and 2017) for assessment of frailty (≥ 3 criteria from the frailty phenotype: low physical activity, slow walking speed, muscle weakness, weight loss and exhaustion). All‐cause mortality was assessed up to January 2022. Analyses were conducted using Cox proportional hazard models and multistate models adjusted for sociodemographic, lifestyle and other dietary factors. Results Mean consumption of protein was 90.2 (standard deviation [SD]: 26.8) g/day, which represents 18.7% of the total energy intake and 1.23 (0.39) g per kg of body weight per day. Plant protein represented 6.16% of the energy intake, while animal protein represented 12.5%. During a median follow‐up of 12.9 (range: 11.7–13.9) years, we documented 196 incident cases of frailty and 490 deaths; of these mortality cases, 83 individuals died after a frailty diagnosis. Higher intake of total protein was associated with decreased risk of frailty (hazard ratio [HR] for tertile 3 vs. tertile 1: 0.66; 95% confidence interval [CI]: 0.45, 0.96; p trend: 0.03). In multistate models, higher fish protein intake decreased the risk in the progression from multimorbidity to frailty (HR per 1‐SD increment: 0.81 [95% CI: 0.68, 0.97]), and higher plant protein decreased the risk of progressing from multimorbidity to mortality (0.86 [0.75, 0.98]). In the progression from frailty to mortality, estimates for total, plant and animal protein showed increased risk (HR for 1 SD increment in total protein: 1.38 [1.05, 1.81]; HR for plant protein: 1.29 [1.01, 1.67]; HR for animal protein: 1.41 [1.04, 1.92]). No significant associations were found between meat protein and dairy protein in any transition. Conclusions In individuals with multimorbidity, higher protein intake, especially fish protein, was associated with lower risk of subsequent frailty, whereas plant protein intake was associated with lower risk of mortality. Higher total protein intake, however, might be detrimental in patients with multimorbidity and frailty. Trial Registration ClinicalTrials.gov identifier: NCT02804672.

Eleni Straitouri, Stratis Tsirtsis, Ander Artola Velasco et al.

Recent work has shown that, in classification tasks, it is possible to design decision support systems that do not require human experts to understand when to cede agency to a classifier or when to exercise their own agency to achieve complementarity$\unicode{x2014}$experts using these systems make more accurate predictions than those made by the experts or the classifier alone. The key principle underpinning these systems reduces to adaptively controlling the level of human agency, by design. Can we use the same principle to achieve complementarity in sequential decision making tasks? In this paper, we answer this question affirmatively. We develop a decision support system that uses a pre-trained AI agent to narrow down the set of actions a human can take to a subset, and then asks the human to take an action from this action set. Along the way, we also introduce a bandit algorithm that leverages the smoothness properties of the action sets provided by our system to efficiently optimize the level of human agency. To evaluate our decision support system, we conduct a large-scale human subject study ($n = 1{,}600$) where participants play a wildfire mitigation game. We find that participants who play the game supported by our system outperform those who play on their own by $\sim$$30$% and the AI agent used by our system by $>$$2$%, even though the AI agent largely outperforms participants playing without support. We have made available the data gathered in our human subject study as well as an open source implementation of our system at https://github.com/Networks-Learning/narrowing-action-choices .

He Zhang, Xinyi Fu, John M. Carroll

Traditional image annotation tasks rely heavily on human effort for object selection and label assignment, making the process time-consuming and prone to decreased efficiency as annotators experience fatigue after extensive work. This paper introduces a novel framework that leverages the visual understanding capabilities of large multimodal models (LMMs), particularly GPT, to assist annotation workflows. In our proposed approach, human annotators focus on selecting objects via bounding boxes, while the LMM autonomously generates relevant labels. This human-AI collaborative framework enhances annotation efficiency by reducing the cognitive and time burden on human annotators. By analyzing the system's performance across various types of annotation tasks, we demonstrate its ability to generalize to tasks such as object recognition, scene description, and fine-grained categorization. Our proposed framework highlights the potential of this approach to redefine annotation workflows, offering a scalable and efficient solution for large-scale data labeling in computer vision. Finally, we discuss how integrating LMMs into the annotation pipeline can advance bidirectional human-AI alignment, as well as the challenges of alleviating the "endless annotation" burden in the face of information overload by shifting some of the work to AI.

Paula Akemi Aoyagui, Kelsey Stemmler, Sharon Ferguson et al.

In subjective decision-making, where decisions are based on contextual interpretation, Large Language Models (LLMs) can be integrated to present users with additional rationales to consider. The diversity of these rationales is mediated by the ability to consider the perspectives of different social actors. However, it remains unclear whether and how models differ in the distribution of perspectives they provide. We compare the perspectives taken by humans and different LLMs when assessing subtle sexism scenarios. We show that these perspectives can be classified within a finite set (perpetrator, victim, decision-maker), consistently present in argumentations produced by humans and LLMs, but in different distributions and combinations, demonstrating differences and similarities with human responses, and between models. We argue for the need to systematically evaluate LLMs' perspective-taking to identify the most suitable models for a given decision-making task. We discuss the implications for model evaluation.

Vasanth Reddy Baddam, Behdad Chalaki, Vaishnav Tadiparthi et al.

In social robot navigation, traditional metrics like proxemics and behavior naturalness emphasize human comfort and adherence to social norms but often fail to capture an agent's autonomy and adaptability in dynamic environments. This paper introduces human empowerment, an information-theoretic concept that measures a human's ability to influence their future states and observe those changes, as a complementary metric for evaluating social compliance. This metric reveals how robot navigation policies can indirectly impact human empowerment. We present a framework that integrates human empowerment into the evaluation of social performance in navigation tasks. Through numerical simulations, we demonstrate that human empowerment as a metric not only aligns with intuitive social behavior, but also shows statistically significant differences across various robot navigation policies. These results provide a deeper understanding of how different policies affect social compliance, highlighting the potential of human empowerment as a complementary metric for future research in social navigation.

Kate Letheren, Nicole Robinson

Humans and robots are increasingly working in personal and professional settings. In workplace settings, humans and robots may work together as colleagues, potentially leading to social expectations, or violation thereof. Extant research has primarily sought to understand social interactions and expectations in personal rather than professional settings, and none of these studies have examined negative outcomes arising from violations of social expectations. This paper reports the results of a 2x3 online experiment that used a unique first-person perspective video to immerse participants in a collaborative workplace setting. The results are nuanced and reveal that while robots are expected to act in accordance with social expectations despite human behavior, there are benefits for robots perceived as being the bigger person in the face of human rudeness. Theoretical and practical implications are provided which discuss the import of these findings for the design of social robots.

Anthony J. Weinhaus

Luka Pušnik, Lisa Lechner, Igor Serša et al.

Abstract Background The complex anatomy of peripheral nerves has been traditionally investigated through histological microsections, with inherent limitations. We aimed to compare three-dimensional (3D) reconstructions of median and ulnar nerves acquired with tomographic high-resolution ultrasound (HRUS) and magnetic resonance microscopy (MRM) and assess their capacity to depict intraneural anatomy. Methods Three fresh-frozen human upper extremity specimens were prepared for HRUS imaging by submersion in a water medium. The median and ulnar nerves were pierced with sutures to improve orientation during imaging. Peripheral nerve 3D HRUS scanning was performed on the mid-upper arm using a broadband linear probe (10–22 MHz) equipped with a tomographic 3D HRUS system. Following excision, nerves were cut into 16-mm segments and loaded into the MRM probe of a 9.4-T system (scanning time 27 h). Fascicle and nerve counting was performed to estimate the nerve volume, fascicle volume, fascicle count, and number of interfascicular connections. HRUS reconstructions employed artificial intelligence-based algorithms, while MRM reconstructions were generated using an open-source imaging software 3D slicer. Results Compared to MRM, 3D HRUS underestimated nerve volume by up to 22% and volume of all fascicles by up to 11%. Additionally, 3D HRUS depicted 6–60% fewer fascicles compared to MRM and visualized approximately half as many interfascicular connections. Conclusion MRM demonstrated a more detailed fascicular depiction compared to 3D HRUS, with a greater capacity for visualizing smaller fascicles. While 3D HRUS reconstructions can offer supplementary data in peripheral nerve assessment, their limitations in depicting interfascicular connections and small fascicles within clusters necessitate cautious interpretation. Clinical relevance statement Although 3D HRUS reconstructions can offer supplementary data in peripheral nerve assessment, even in intraoperative settings, their limitations in depicting interfascicular branches and small fascicles within clusters require cautious interpretation. Key Points 3D HRUS was limited in visualizing nerve interfascicular connections. MRM demonstrated better nerve fascicle depiction than 3D HRUS. MRM depicted more nerve interfascicular connections than 3D HRUS. Graphical Abstract

Sabrina Vieu, Ninon Le Poul, Léa Tur et al.

Abstract Axolotls (Ambystoma mexicanum) are extensively studied for their relevance in human medical research. Despite being critically endangered in the wild, they have gained popularity as household pets. Although they have been kept in captivity for over a century, detailed descriptions of their coelomic organ anatomy remain limited. Also, this species exhibits significant variations compared to other amphibians. Ultrasound is a non-invasive and painless medical imaging technique, ideally suited for investigating internal organs or structures. This study focused on describing the ultrasound appearance of the axolotl coelomic cavity. It details the identification, localization and parenchymal description of major organs in 28 neotenic axolotls using ultrasound frequencies ranging from 7 to 15 MHz. The accuracy of the results was validated by comparing ultrasound findings with necropsy results from one male and one female axolotl. The heart, lung surface, liver and reproductive tracts were visualized. Measurements, along with confidence intervals, were calculated for the spleen, kidneys, testicles, gastric wall, gallbladder, and pylorus. Occasional detection of hyperechoic millimetric particles in the gallbladder or ascites was noted. However, visualization of the pancreas and bladder was not possible. This research outcomes involve the development of a comprehensive atlas comprising images obtained throughout the study. Additionally, the experiment established a reproducible and readily accessible protocol for conducting anatomy-morphological assessments in axolotl medicine. This protocol stands as a crucial preliminary stage before advancing to lesion identification.

Hye‐Sun Park, Sang Ho Park, Yeseul Seong et al.

Abstract Background The association between the adiponectin‐to‐leptin ratio (A/L ratio) and the risk of incident chronic kidney disease (CKD) is poorly understood. This study aimed to investigate the association between A/L ratio and the risk of incident CKD and to examine whether such a relationship varied according to sex and body composition. Methods In this prospective community‐based cohort, participants with normal kidney function were analysed (N = 5192). The association between the A/L ratio at baseline and the risk of incident CKD, defined as two or more occasions with an estimated glomerular filtration rate of <60 mL/min/m2 or proteinuria of ≥1+ on a dipstick test during the follow‐up period, was evaluated using multivariable Cox proportional hazards analyses. Subgroup analyses were conducted based on sex, body mass index (BMI) and the presence of sarcopenia. Results The participants' mean age was 57.2 ± 8.3 years, and 53.2% were women. The A/L ratio was higher in men compared with women (1.5 [0.8–3.2] and 0.5 [0.3–0.9] μg/ng, P < 0.001). During a median follow‐up of 9.8 [9.5–10.0] years, 417 incident CKD events occurred (8.7 per 1000 person‐years). Men in the highest quartile of A/L ratio had a lower risk of incident CKD (adjusted hazard ratio [aHR], 0.57; 95% confidence interval [CI], 0.33–0.99) than those in the lowest quartile. Additionally, a 1.0 increase in A/L ratio was associated with a 12% decreased risk of incident CKD in men (aHR, 0.88; 95% CI, 0.80–0.97). However, no significant association was observed in women. In subgroup analysis stratified by BMI and the presence of sarcopenia, the association between a high A/L ratio and a reduced risk of incident CKD was consistent in men with a BMI < 23.0 kg/m2 and those with sarcopenia. However, no significant association was observed between men with a BMI ≥ 23.0 kg/m2 and those without sarcopenia. Conclusions A high A/L ratio is an independent marker of a reduced risk of incident CKD in men, especially in those with a BMI < 23.0 kg/m2 and sarcopenia.

Erina Seh-Young Moon, Shion Guha

Homelessness is a humanitarian challenge affecting an estimated 1.6 billion people worldwide. In the face of rising homeless populations in developed nations and a strain on social services, government agencies are increasingly adopting data-driven models to determine one's risk of experiencing homelessness and assigning scarce resources to those in need. We conducted a systematic literature review of 57 papers to understand the evolution of these decision-making algorithms. We investigated trends in computational methods, predictor variables, and target outcomes used to develop the models using a human-centered lens and found that only 9 papers (15.7%) investigated model fairness and bias. We uncovered tensions between explainability and ecological validity wherein predictive risk models (53.4%) focused on reductive explainability while resource allocation models (25.9%) were dependent on unrealistic assumptions and simulated data that are not useful in practice. Further, we discuss research challenges and opportunities for developing human-centered algorithms in this area.

Paula Akemi Aoyagui, Sharon Ferguson, Anastasia Kuzminykh

An essential aspect of evaluating Large Language Models (LLMs) is identifying potential biases. This is especially relevant considering the substantial evidence that LLMs can replicate human social biases in their text outputs and further influence stakeholders, potentially amplifying harm to already marginalized individuals and communities. Therefore, recent efforts in bias detection invested in automated benchmarks and objective metrics such as accuracy (i.e., an LLMs output is compared against a predefined ground truth). Nonetheless, social biases can be nuanced, oftentimes subjective and context-dependent, where a situation is open to interpretation and there is no ground truth. While these situations can be difficult for automated evaluation systems to identify, human evaluators could potentially pick up on these nuances. In this paper, we discuss the role of human evaluation and subjective interpretation to augment automated processes when identifying biases in LLMs as part of a human-centred approach to evaluate these models.

David Helmer, Michael Boardman, S. Kate Conroy et al.

The REAIM 2024 Blueprint for Action states that AI applications in the military domain should be ethical and human-centric and that humans must remain responsible and accountable for their use and effects. Developing rigorous test and evaluation, verification and validation (TEVV) frameworks will contribute to robust oversight mechanisms. TEVV in the development and deployment of AI systems needs to involve human users throughout the lifecycle. Traditional human-centred test and evaluation methods from human factors need to be adapted for deployed AI systems that require ongoing monitoring and evaluation. The language around AI-enabled systems should be shifted to inclusion of the human(s) as a component of the system. Standards and requirements supporting this adjusted definition are needed, as are metrics and means to evaluate them. The need for dialogue between technologists and policymakers on human-centred TEVV will be evergreen, but dialogue needs to be initiated with an objective in mind for it to be productive. Development of TEVV throughout system lifecycle is critical to support this evolution including the issue of human scalability and impact on scale of achievable testing. Communication between technical and non technical communities must be improved to ensure operators and policy-makers understand risk assumed by system use and to better inform research and development. Test and evaluation in support of responsible AI deployment must include the effect of the human to reflect operationally realised system performance. Means of communicating the results of TEVV to those using and making decisions regarding the use of AI based systems will be key in informing risk based decisions regarding use.