AI Levels Research Articles

Can I Collaborate with You? An Investigation of Trust in Generative AI

ABSTRACTThis study investigates the role of trust in human‐AI collaboration in K‐12 education, focusing on generative AI. Using a two‐factor experimental design (n = 26), we examined attitudes toward AI and creativity level as independent variables and assessed cognitive trust, affective trust, perceptions of usability, and continuance intention to use generative AI tools in text‐to‐text (ChatGPT) and text‐to‐image (Canva) tasks. The results is analyzed with two‐way ANOVA, we find that: (1) text‐to‐text human‐AI collaboration tends to foster higher emotional engagement and trust compared to text‐to‐image human‐AI collaboration; (2) perceptions of usability and continuance intention to use are slightly higher in text‐to‐text interactions, suggesting a preference for text‐based generative AI tools; (3) significant interactive effects between attitudes toward AI and creativity were observed across all measured variables, highlighting the importance of considering both factors in understanding and improving human‐AI collaboration in educational settings, contributing to research on generative AI for learning assistance.

Bridging the gap: Mechanistic-based cyclist injury risk curves using two decades of crash data

Objective Injury risk curves are vital in quantifying the relative safety consequences of real-world collisions. Previous injury risk curves for bicycle-passenger vehicle crashes have predominantly focused on frontal impacts. This creates a gap in cyclist injury risk assessment for other geometric crash configurations. The goal of this study was to create an “omnidirectional” injury risk model, informed by known injury causing mechanisms, that is applicable to most geometric configurations. Methods We used data from years 1999–2022 of the German In-Depth Accident Study (GIDAS). We describe the pattern of injuries for cyclists involved in collisions with passenger vehicles, and we developed injury risk functions at various AIS levels for these collisions. A mechanistic-based approach accounting for biomechanically-relevant variables was used to select model parameters a priori. Cyclist age (including children) and sex were regarded as relevant predictors of injury risk. Speed and impact geometry were captured through a novel predictor, Effective Collision Speed, which transforms the vehicle and cyclist speeds into a single value and incorporates frictional considerations observed during side engagements. Cyclist engagement with the vehicle was captured with a variable demonstrating the potential for a normal projection. We additionally present analyses weighted toward German nationwide data. Results We identified 6,576 cyclists involved in collisions with passenger vehicles. AIS3+ cyclist injuries occurred most often in the head, thorax, and lower extremities. Effective Collision Speed was a strong predictor of injury risk. Collisions with a potential for a normal projection were associated with increased risk, though this was only significant at the MAIS2+F severity level. Younger children had slightly higher injury risk compared to young adults, while elderly cyclists had the highest risk of AIS3+ injury. Sex was a significant predictor only for the MAIS2+F injury risk curves. Significance U.S. cyclist fatalities increased 55% from 2010 to 2021. To reduce injuries and fatalities, it is crucial to understand cyclist injury risk. This study builds on previous analyses by including children, incorporating additional mechanistic predictors, broadening the scope of included crashes, and using weighting to generalize these estimates toward national German statistics.

Quantifying Treatments as Usual and with Technologies in Neurorehabilitation of Individuals with Spinal Cord Injury.

Several technologies have been introduced into neurorehabilitation programs to enhance traditional treatment of individuals with Spinal Cord Injury (SCI). Their effectiveness has been widely investigated, but their adoption has not been properly quantified. The aim of this study is to assess the distribution of conventional (Treatment As Usual-TAU) and technology-aided (Treatment With Technologies-TWT) treatments conveniently grouped based on different therapeutic goals in a selected SCI unit. Data from 104 individuals collected in 29 months were collected in a custom database and categorized according to both the conventional American Impairment Scale classification and a newly developed Multifactor (MF) clustering approach that considers additional sources of information (the lesion level, the level of independence in the activities of daily living, and the hospitalization duration). Results indicated an average technology adoption of about 30%. Moreover, the MF clusters were less overlapped, and the differences in TWT adoption were more pronounced than in AIS-based clustering. MF clustering was capable of grouping individuals based both on neurological features and functional abilities. In particular, individuals with motor complete injuries were grouped together, whereas individuals with sensorimotor incomplete SCI were collected separately based on the lesion level. As regards TWT adoption, we found that in the case of motor complete SCI, TWT for muscle tone control and modulation was mainly selected (about 90% of TWT), while the other types of TWT were seldom adopted. Even for individuals with incomplete SCI, the most frequent rehabilitation goal was muscle tone modulation (about 75% of TWT), regardless of the AIS level, and technologies to improve walking ability (about 12% of TWT) and balance control (about 10% of TWT) were mainly used for individuals with thoracic or lumbar lesions. Analyzing TAU distribution, we found that the highest adoption of muscle tone modulation strategies was reported in the case of individuals with motor complete SCI (about 42% of TAU), that is, in cases when almost no gait training was pursued (about 1% of TAU). In the case of cervical motor incomplete SCI, compared to thoracic and lumbar incomplete SCI, there was a greater focus on muscle tone control and force recruitment in addition to walking training (38% and 14% of TAU, respectively) than on balance training. Overall, the MF clustering provided more insights than the traditional AIS-based classification, highlighting differences in TWT adoption. These findings suggest that a wider overview that considers both neurological and functional characteristics of individuals after SCI based on a multifactor analysis could enhance the personalization of neurorehabilitation strategies.

Metabolic Consequences of Polycystic Ovary Syndrome on Hepatic Function in High Fat Diet-Fed Rats: Potential Role of Exercise

Abstract Context Polycystic ovarian syndrome (PCOS) is the most common endocrine-metabolic disorder in reproductive-aged women. Since, The current study was designed to elucidate the metabolic comorbidities of polycystic ovarian syndrome, especially the hepatic effect, with or without high fat diet feeding, and to emphases the crucial role of exercise in managing such complications. Materials and Methods This study was carried out on 55 adult female albino Wistar rats, the study duration was 8 weeks. Rats included in the present study were randomly allocated into five equal groups, eleven rats each: Group I: Control group, Group II: Polycystic ovary syndrome (PCO) group, Group III: High fat diet fed polycystic ovary syndrome (HF-PCO) group, Group IV: Exercise-treated polycystic ovary syndrome (PCO-EX) group, Group V: High fat diet fed exercise-treated polycystic ovary syndrome (HF-PCO-EX) group. PCOS was induced in female wistar rats by oral administration of Letrozole (1mg/kg daily by oral gavage for 21 days in groups II, III, IV, V, high fat diet was supplemented to groups III and V. Rats of groups IV and V were subjected to 30 minutes swimming protocol 5 days/ week for 5 weeks after the period of induction. Results All measured values of the OGTT as well as area under the curve, serum fasting glucose (FG), insulin levels and HOMA-IR were all significantly increased in the PCO, HF-PCO when compared to control group. Compared to their corresponding non exercised groups, both PCO-EX and HFPCO-EX groups showed significant reductions in all parameters except for T90 in PCO-EX group. For the lipid profile, the serum level of TG, TC, LDL-C and AI were all significantly increased while HDL-C was significantly decreased in PCO and HF-PCO compared to control group as well as in HF-PCO group compared to PCO group. PCO-EX and HF-PCO-EX groups showed significant reductions in TG, TC, LDL-C and AI and a significant elevation in HDL-C when compared to their corresponding non- exercised groups. For liver enzymes serum level of ALT, AST and GGT were significantly increased in PCO and HF-PCO groups when compared with control group. Compared to their corresponding non exercised groups, both PCO-EX and HF-PCO-EX showed significant reductions in liver enzymes. As regard the hormonal profile, the serum level of estradiol, adiponectin and adiponectin/resistin ratio (A/R) were significantly decreased while serum level of testosterone, LH and resistin were significantly increased in PCO and HF-PCO groups compared to control group. PCO-EX and HF-PCO-EX groups induced significant increases in serum estradiol, adiponectin and A/R as well as significant decreases in serum testosterone, LH and resistin compared to their corresponding non exercised groups. Conclusion We could conclude that both moderate intensity exercise and avoiding high fat diet have beneficial therapeutic potentials in ameliorating the metabolic effects of PCO syndrome in rats especially the hepatic effects and the development of NAFLD.

Insight from industry-early lessons learned about AI adoption in core procurement processes, directions for managers and researchers

PurposeThere is a growing body of conceptual work considering the potential of AI in supply chain and procurement, and there is great interest in AI among managers. But, according to a recent study, digital strategies for procurement are often missing or not satisfactory. Literature offers conflicting guidance on possible adoption areas for AI in core procurement processes. Given the need for better digital strategies for procurement and the need to further develop the understanding of adoption potential, the purpose of this paper is to explore actual adoption levels, experienced benefits, readiness levels and barriers to implementation in industry. This informs nuanced, not hyped, managerial consideration and identifies further research opportunities.Design/methodology/approachLeveraging items used in literature to study adoption of other technologies, the authors conduct the first empirical exploration of actual adoption levels of AI in procurement. The authors do so by collecting survey responses in three manager workshops, and the authors use the workshops to seek manager input in the interpretation of findings and the identification of implications for managers and researchers.FindingsThere appears to be less consideration given to AI in procurement than interest in the topic might imply. Adoption levels are generally low, implying that there is a lot of room for the development of consideration, use cases and possible pilots by managers and researchers. The authors find procurement benefits of AI adoption to be broader than costs and productivity alone, including visibility and innovation. But, readiness appears to be at relatively low levels with factors commonly considered in literature, such as executive support and willingness to invest, less relevant than less widely considered elements such as human sense making and supplier readiness.Originality/valueThis first empirical exploration moves past conceptualization and the study of potential adoption into the study of actual adoption levels in different procurement core processes. The authors expand the consideration of readiness by including additional items of human sense making as called for in literature. The authors also include and develop supplier readiness consideration, which is often missing from research. With the help of participating managers, the authors are able to develop a more comprehensive framework for the consideration of AI adoption. This can help bring nuance, not hype, to consideration and provides a rich portfolio of research items and constructs to further explore.

The dark side of artificial intelligence in marketing: meta-analytics review

PurposeArtificial intelligence (AI) has become a pivotal technology in both marketing and daily life. Despite extensive research on the benefits of AI, its adverse effects on customers have received limited attention.Design/methodology/approachWe employed meta-analysis to synthesise effect sizes from 45 studies encompassing 50 independent samples (N = 19,503) to illuminate the negative facets of AI's impact on customer responses.FindingsAdverse effects of AI, including privacy concern, perceived risks, customer alienation, and uniqueness neglect, have a negative and significant effect on customers' cognitive (perceived benefit, trust), affective (attitude and satisfaction) and behavioural responses (purchase, loyalty, well-being). Additionally, moderators in AI (online versus offline), customer (age, male vs. female), product (hedonic vs. utilitarian, high vs. low involvement), and firm level (service vs. manufacturing) and national level (individualism, power distance, masculinity, uncertainty avoidance, long-term orientation) moderate these relationships.Practical implicationsOur findings inform marketing managers about the drawbacks of utilising AI as part of their value proposition and provide recommendations on how to minimise these effects in different contexts. Additionally, policymakers need to consider the dark side of AI, especially among the vulnerable groups.Originality/valueThis paper is among the first research studies that synthesise previous research on the dark side of AI, providing a comprehensive view of its diminishing impact on customer responses.

A Randomized, Double-Blind, Placebo-Controlled Clinical Trial on the Effect of a Dietary Supplement Containing Dry Artichoke and Bergamot Extracts on Metabolic and Vascular Risk Factors in Individuals with Suboptimal Cholesterol Levels.

The aim of this study was to assess whether dietary supplementation with a nutraceutical blend comprising extracts of bergamot and artichoke-both standardized in their characteristic polyphenolic fractions-could positively affect serum lipid concentration and insulin sensitivity, high-sensitivity C-reactive protein (hs-CRP), and indexes of non-alcoholic fatty liver disease (NAFLD) in 90 healthy individuals with suboptimal cholesterol levels. Participants were randomly allocated to treatment with a pill of either active treatment or placebo. After 6 weeks, the active-treated group experienced significant improvements in levels of triglycerides (TG), apolipoprotein B-100 (Apo B-100), and apolipoprotein AI (Apo AI) versus baseline. Total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), non-high density lipoprotein cholesterol (Non-HDL-C), and hs-CRP also significantly decreased in the active-treated group compared to both baseline and placebo. At the 12-week follow-up, individuals allocated to the combined nutraceutical experienced a significant improvement in TC, LDL-C, Non-HDL-C, TG, Apo B-100, Apo AI, glucose, alanine transaminase (ALT), gamma-glutamyl transferase (gGT), hs-CRP, several indexes of NAFLD, and brachial pulse volume (PV) in comparison with baseline. Improvements in TC, LDL-C, Non-HDL-C, TG, fatty liver index (FLI), hs-CRP, and endothelial reactivity were also detected compared to placebo (p < 0.05 for all). Overall, these findings support the use of the tested dietary supplement containing dry extracts of bergamot and artichoke as a safe and effective approach for the prevention and management of a broad spectrum of cardiometabolic disorders.

Investigating the impact of artificial intelligence development on water pollution in China

Analysing the impacts of AI development on water pollution is of great significance for improving water environment management measures and promoting global sustainable development. In this study, macro and micro panel data of China from 2008 to 2019 were used to measure the development level of AI at the industry and provincial levels in China, and a panel data regression model was constructed to analyse the effects of AI development on water pollution and regional heterogeneity. The measurement results show that the development level of AI is increasing. Among the six industries, the intelligence level of the manufacturing industry has developed the fastest, by approximately 48.74 times compared with the baseline. At the provincial level, Guangdong Province made the greatest progress in intelligence level, increasing by approximately 23.6 times. The regression results show that AI technology is helpful in reducing the level of water pollution, and this conclusion remains valid after the regression of instrumental variables and robustness testing. In terms of regional heterogeneity, AI technology can significantly reduce the degree of water pollution in eastern and western China. However, the development of AI technology in the central region has no significant impact on water pollution, and a risk of increasing the degree of water pollution exists. The reasons may be that the eastern region has high levels of economic vitality and technological innovation and a strong ability to transform AI achievements, which can significantly improve the water environment. Industrial support policies in the western region are strong, and the application of AI can quickly improve production methods and significantly reduce water pollution levels. In comparison, the levels of economic development, technological innovation, and policy support in the central region need to be strengthened, so the environmental effect of AI is not significant.

The Applications of Artificial Intelligence in Human Resources

The study examines the transformative role of artificial intelligence in Human resources and its profound implications for the future of workforce management. Questionnaires are used to evaluate the current perception, applications &amp; adoption of artificial intelligence in human resources practices, as well as its potential impact and ability to shape tomorrow's Human Resource landscape. Understanding of how people view the integration of AI in human resources functions, by rigorously analysing questionnaire responses. These findings reveal the apprehensions, expectations and acceptance levels of AI driven human resource management processes. In addition, important foresight is provided on the evolution of technology and Human Capital dynamics within organisational structures. Moreover, a compelling case study is presented to demonstrate the effectiveness of AI when it comes to human resource management on an actual basis. The case study shows how AI solutions simplify recruitment, enhance employee engagement, optimise talent management and mitigate biases in the decision making process for a more agile, data driven environment that is inclusive. Finally, this research aims at breaking down the complex implications of artificial intelligence for human resources and providing information on its transformative potential which will make it easier to adopt effective decisions as part of organisational HR strategies. Keywords: Artificial Intelligence, Human Resources, Case Study, Perception, Awareness

Serum CircNIPSNAP3A is Associated with Metabolic Disorders, Atherosclerosis and Severity of Coronary Artery Disease in a Chinese Population.

The relationships of serum circNIPSNAP3A and circHIPK3 with metabolic disorders, atherosclerosis and severity of coronary artery disease (CAD) remain to be clarified. Three hundred and thirty-eight subjects were categorized into normal coronary artery, atherosclerosis and CAD groups. Clinical data including anthropometric indexes, medical history, and physiological and biochemical parameters were collected. Serum circNIPSNAP3A and circHIPK3 were determined by quantitative real-time PCR. CAD severity was evaluated by clinical manifestation, electrocardiogram and coronary angiography. Both CAD and atherosclerosis groups had a higher serum level of circNIPSNAP3A than the normal coronary artery group (P < 0.05 for all). The subjects with a high percentage (> 66th percentile) of circNIPSNAP3A had higher mean levels of triglycerides, uric acid and homocysteine, and lower mean levels of high-density lipoprotein cholesterol and apolipoprotein AI than those with a low percentage (< 33rd percentile) of circNIPSNAP3A. Notably, circNIPSNAP3A is significantly and independently associated with CAD, and subjects with a high percentage of circNIPSNAP3A had more diseased coronary branches and a higher incidence of acute coronary syndrome than those with a low percentage of circNIPSNAP3A. Regarding circHIPK3, subjects with a medium or high percentage of circHIPK3 had a lower mean level of apolipoprotein AI than those with a low percentage of circHIPK3, but no significant differences in the incidence and severity of CAD among the < 33rd, 33rd-66th, and > 66th percentiles of circHIPK3 were detected. Serum circNIPSNAP3A is related to cardiovascular risk factors and CAD severity, and may be a potential prognostic marker and/or therapeutic target for CAD.

Artificial Intelligence: The Future

Artificial intelligence is the intelligence of machines or software, as opposed to the intelligence of humans or animals. It is also the field of study in computer science that develops and studies intelligent machines. "AI" may also refer to the machines themselves. AI is not a new for the scientist, it was introduce in 1943 with artificial neurons model and get popular in 1950 due to “Turting test” the test was done to get answer that machine can think?, purposed by Alan Turing. Basically AI is categorized into three types, Artificial Narrow Intelligence, Artificial General Intelligence and Artificial Super Intelligence. Deep learning and machine learning is major subfield of Artificial intelligence. DL as a subset of ML, which is also another subset of AI. Therefore, AI is the all-encompassing concept that initially erupted. Application of AI fields are Healthcare, Business, Education, Agriculture, Finance, Law, Entertainment and media, Software coding and IT processes, Security, Manufacturing, Banking and Transportation. In reference to Job creation or distraction, Artificial Intelligence is not job killer but a job category killer”. In the latest report (May 2023) on The Future of Jobs, the World Economic Forum (WEF) predicts the creation of 69 million jobs by 2027 thanks to AI, but also the destruction of 89 million jobs. In context to intelligence level, Jan. 2022, Age of IQ level of AI was as 7 year Children and in Dec. 2022, Age of IQ level of AI was as 9 year Children. India is emerging market in global and it has around 12 % of work could be automated by AI. In India more than 2000 startup are related to AI and 90000 plus AI Professional work in India.The economic impact of AI, for select G20 countries and estimates AI to boost India’s annual growth rate by 1.3 percentage points by 2035. AI has potential to add 1 trillion to India’s economy in 2035. We are going to enter into new technological world, it’s may be our fortune or misfortune. Key Words: - Artificial Intelligence, Machine learning, Job, India

Artificial Intelligence Anxiety of Nurses and Related Factors

The research was carried out to examine the artificial intelligence anxiety levels of nurses and their affecting factors. In this study, a descriptive and cross-sectional design was used. The sample of the study consisted of 120 nurses (n=120). The research data were collected between 10 July and 10 October 2021. Data were collected using a Nurse Information From (NIF) and Artificial Intelligence Anxiety Scale (AIAS). Mann- Whitney U and Kruskal-Wallis test were used in the analysis of research data. The average age of nursing was 31.05±7.40 and 82.2% were females. The AI levels of the nurses were found to be 43.36±11.13. It was determined that there was a difference between the educational status of the nurses, their knowledge of AI technologies, the effect of AI technologies in patient care and their AI anxiety levels (p

The promise of AI in medicine: A call for greater education for Pakistani medical professionals.

Madam, AI is revolutionizing the world in all spheres and the institution of medicine worldwide is no exception. While we are most familiar with its contribution to radiology, AI has now extensive applications within dermatology, ophthalmology, pathology, medicine, ophthalmology, neurosciences and more. Through progressive efforts of both doctors and programmers, AI is now becoming more adept in the field of medicine. Some of its applications in modern medicine include staging of skin cancers by looking at dermatoscopic images (1), classifying age-related macular degeneration and diabetic macular edema by through OCT images (2), and more. The accuracy by which AI performs these tasks is comparable to a human being. However, Pakistan being in the developing stage is still lagging behind. With inadequate resources, constant economic desolation, and limited knowledge and data in the field of IT, the country is far behind in this rapidly progressive field. According to a study that included participants throughout Pakistan 74% of doctors and 68.8% of medical students had a basic knowledge of AI but only 27.3% of doctors and 19.4% of students were aware of its medical applications (3). In another study only 56.7% of the participants were familiar with AI and its subtypes, which highlighted the unawareness about AI in more than 40% of the participants. These studies clearly show the deficit in knowledge that medical personnel have about AI, its applications and impact on health care. AI and its rapid progress in health sectors worldwide are creating discrepancies between them and us. Hence, the senior authorities in our education system should come up with a proposition that can shorten this gap. By introducing basic levels of AI in the medical curriculum with a course spanning only around 1 week, including basics of coding and computational thinking (5), our medical personnel will be able to incorporate a fundamental level of computational thinking and coding language. This will help them in adapting to the revolutionized healthcare system as well as give them enough command to grapple with it and give feedback to improvise AI even more. And since medical science is based on logical reasoning and problem-solving, much like IT, we believe that our medical students can and will be able to inculcate this into their skills. This can be our very first step towards a much more efficient and progressive healthcare system.

Customer journey optimisation using large language models: Best practices and pitfalls in generative AI

Today's business environment is moving faster than ever, and the expressive and adaptive capabilities of generative AI (GenAI) and large language models (LLMs) are redefining the enterprise rails of tomorrow. Given the abundance of industry hype, investor expectations and leadership pressure, the initial impulse is to ‘get in the game’. But how does one implement initiatives that drive business outcomes within ethical parameters while avoiding technical pitfalls? Marketers need practical guidance to navigate through these changes. In this paper, the authors examine multiple considerations for deployment of GenAI in marketing and customer experience. How does the marketer decide on which initiatives and opportunities to begin with? Which use cases will drive value as the organisation adapts to deploying these new capabilities? Once a marketer has identified the opportunities to capitalise on through GenAI, how is the capability deployed? There are a variety of approaches that can be considered given the level of organisational capability with AI and resource levels to be applied. As with any cutting-edge capability, there are potential missteps that must be avoided to ensure success. This paper provides some insight based on practical experiences to date that cover ethical, technical and process concerns. The paper presents thoughtful approaches to the deployment of LLMs and GenAI that can result in concrete ROI and reduced risk even in this early stage of adoption. With this information, marketers can be prepared to confidently begin their journey using GenAI to transform their customer experience and drive enterprise value for their organisations.

Inhibiting Rev-erbα-mediated ferroptosis alleviates susceptibility to myocardial ischemia-reperfusion injury in type 2 diabetes

The complex progression of type-2 diabetes (T2DM) may result in increased susceptibility to myocardial ischemia-reperfusion (IR) injury. IR injuries in multiple organs involves ferroptosis. Recently, the clock gene Rev-erbα has aroused considerable interest as a novel therapeutic target for metabolic and ischemic heart diseases. Herein, we investigated the roles of Rev-erbα and ferroptosis in myocardial IR injury during T2DM and its potential mechanisms. A T2DM model, myocardial IR and a tissue-specific Rev-erbα−/− mouse in vivo were established, and a high-fat high glucose environment with hypoxia-reoxygenation (HFHG/HR) in H9c2 were also performed. After myocardial IR, glycolipid profiles, creatine kinase-MB, AI, and the expression of Rev-erbα and ferroptosis-related proteins were increased in diabetic rats with impaired cardiac function compared to non-diabetic rats, regardless of the time at which IR was induced. The ferroptosis inhibitor ferrostatin-1 decreased AI in diabetic rats given IR and LPO levels in cells treated with HFHG/HR, as well as the expression of Rev-erbα and ACSL4. The ferroptosis inducer erastin increased AI and LPO levels and ACSL4 expression. Treatment with the circadian regulator nobiletin and genetically targeting Rev-erbα via siRNA or CRISPR/Cas9 technology both protected against severe myocardial injury and decreased Rev-erbα and ACSL4 expression, compared to the respective controls. Taken together, these data suggest that ferroptosis is involved in the susceptibility to myocardial IR injury during T2DM, and that targeting Rev-erbα could alleviate myocardial IR injury by inhibiting ferroptosis.

Acceptance of AI in Semi-Structured Decision-Making Situations Applying the Four-Sides Model of Communication—An Empirical Analysis Focused on Higher Education

This study investigates the impact of generative AI systems like ChatGPT on semi-structured decision-making, specifically in evaluating undergraduate dissertations. We propose using Davis’ technology acceptance model (TAM) and Schulz von Thun’s four-sides communication model to understand human–AI interaction and necessary adaptations for acceptance in dissertation grading. Utilizing an inductive research design, we conducted ten interviews with respondents having varying levels of AI and management expertise, employing four escalating-consequence scenarios mirroring higher education dissertation grading. In all scenarios, the AI functioned as a sender, based on the four-sides model. Findings reveal that technology acceptance for human–AI interaction is adaptive but requires modifications, particularly regarding AI’s transparency. Testing the four-sides model showed support for three sides, with the appeal side receiving negative feedback for AI acceptance as a sender. Respondents struggled to accept the idea of AI, suggesting a grading decision through an appeal. Consequently, transparency about AI’s role emerged as vital. When AI supports instructors transparently, acceptance levels are higher. These results encourage further research on AI as a receiver and the impartiality of AI decision-making without instructor influence. This study emphasizes communication modes in learning-ecosystems, especially in semi-structured decision-making situations with AI as a sender, while highlighting the potential to enhance AI-based decision-making acceptance.

A comparison between clinical decision support system and clinicians in breast cancer

ObjectiveWe are building a clinical decision support system (CSCO AI) for breast cancer patients to improve the efficiency of clinical decision-making. We aimed to assess cancer treatment regimens given by CSCO AI and different levels of clinicians. Methods400 breast cancer patients were screened from the CSCO database. Clinicians with similar levels were randomly assigned one of the volumes (200 cases). CSCO AI was asked to assess all cases. Three reviewers were independently asked to evaluate the regimens from clinicians and CSCO AI. Regimens were masked before evaluation. The primary outcome was the proportion of high-level conformity (HLC). ResultsThe overall concordance between clinicians and CSCO AI was 73.9% (3621/4900). It was 78.8% (2757/3500) in the early-stage, higher than that in the metastatic stage (61.7% [864/1400], p < 0.001). The concordance was 90.7% (635/700) and 56.4% (395/700) in adjuvant radiotherapy and second-line therapy respectively. HLC in CSCO AI was 95.8% (95%CI:94.0%–97.6%), significantly higher than that in clinicians (90.8%, 95%CI:89.8%–91.8%). Considering professions, the HLC of surgeons was 85.9%, lower than that of CSCO AI (OR = 0.25,95%CI: 0.16–0.41). The most significant difference in HLC was in first-line therapy (OR = 0.06, 95%CI:0.01–0.41). When clinicians were divided according to their levels, there was no statistical significance between CSCO AI and higher level clinicians. ConclusionsDecision from CSCO AI for breast cancer was superior than most clinicians did except in second-line therapy. The improvements in process outcomes suggest that CSCO AI can be widely used in clinical practice.

Predictive accuracy of computer-generated C-index nephrometry scores compared with human-generated scores in predicting oncologic and perioperative outcomes.

623 Background: The Centrality index (C-index) score is a validated nephrometry scoring system that requires precise measurements and mathematical calculations of cross sectional imaging. Like other nephrometry scores, its implementation has been slowed by required time investment and interobserver variability. We sought to automate this score on preoperative computerized tomography scans by developing an artificial intelligence-generated C-index score. We then aimed to evaluate its ability to predict meaningful oncologic and perioperative outcomes as compared to human-generated C-index nephrometry scores. Methods: 300 patients with preoperative computerized tomography with early arterial contrast phase were identified from a cohort of 544 consecutive patients undergoing surgical extirpation for suspected renal cancer. A deep neural network approach was used to automatically segment kidneys and tumors, and then programed to generate the measurements and calculate C-index score. Human C-index scores were independently calculated by medical personnel blinded to AI-scores. AI- and Human score agreement was assessed using bivariate linear regression correlation and their predictive abilities for both oncologic and perioperative outcomes were assessed using logistic regression and compared with receiver operating characteristic (ROC) curve analyses with measurements of areas under the curve (AUC). Results: Median age was 60 years (IQE 51–68), and 40% were female. Median tumor size was 4.2 cm and 91.3% had malignant tumors. 27% were high stage, 37% high grade, and 63% underwent partial nephrectomy. There was significant agreement between Human scores and AI-scores on linear regression analysis (R2 = 0.738, p &lt;0.0001). Both AI- and Human generated C-index scores similarly predicted meaningful oncologic outcomes, with lower levels of either C-index score associated with increased risk of malignant histology (H-score p = 0.018, AI score p =0.014) high-grade disease (both p &lt;0.0001), and high stage disease (both p &lt;0.0001). Lower levels of either AI or human generated C-index scores also predicted a radical nephrectomy rather than partial nephrectomy surgical approach (both p &lt;0.0001). AUC measurements (Table) were similar but consistently superior for AI generated C-index scores. Conclusions: Fully automated AI-generated C-index scores are comparable to human-generated C-index scores and predict a wide variety of meaningful patient-centered outcomes. Once validated in additional populations, our results suggest that our AI generated C-index could be delivered automatically from a preoperative CT scan to a clinician and patient at the point of care to aid in decision making. [Table: see text]

Clinical characteristics of severe coronavirus disease 2019 patients with chronic obstructive pulmonary disease.

Coronavirus disease 2019 (COVID-19) in elderly and patients with chronic respiratory diseases (COPD) had a poor prognosis. COPD is one of the most common chronic respiratory diseases. We explore the epidemiological characteristics of patients with severe COVID-19 with COPD patients in order to provide medical evidence for the prevention and treatment of severe COVID-19. We retrospectively analyzed the clinical baseline characteristics, treatment strategies, disease progression and prognosis of 557 severe COVID-19 patients admitted to the West Court of Union Hospital of Huazhong University of Science and Technology from January 29, 2020 to April 8, 2020. A total of 465 patients with severe COVID-19 were enrolled in the study, including 248 (53.3%) males and 217 (46.7%) females. The median age of severe COVID-19 patients was 62.0 years, and 53 patients were complicated with COPD. Common symptoms at the onset included fever (78.5%), dry cough (67.1%), shortness of breath (47.3%) and fatigue (40.9%). Compared with non-COPD patients, patients with COPD had significantly lower levels of SpO2 in admission (90.0% vs 92.0%, P=0.014). In terms of laboratory examinations, patients with COPD had higher levels of C-reactive protein, interleukin-6, procalcitonin, total bilirubin, blood urea nitrogen, serum creatinine, lipoprotein (a), high-sensitivity troponin I, and D-dimer, while had lower levels of platelet counts, albumin and apolipoprotein AI. Severe COVID-19 patients with COPD had higher Sequential Organ Failure Assessment scores [3.0(2.0, 3.0) vs 2.0(2.0, 3.0), P=0.038] and CURB-65 score [1.0(1.0, 2.0) vs1.0(0.0, 1.0), P<0.001], and a higher proportion of progressing to critical illness (28.3% vs 10.0%, P<0.001) with more complications [e.g. septic shock (15.1% vs 6.1%, P=0.034)], had higher incidence rates of antibiotic therapies (90.6% vs 77.2%, P=0.025), non-invasive (11.3% vs 1.7%, P<0.001) and invasive mechanical ventilation (17.0% vs 8.3%, P=0.039), ICU admission (17.0% vs 7.5%, P=0.021) and death (15.1% vs 6.1%, P=0.016). Cox proportion hazard model was carried out, and the results showed that comorbid COPD was an independent risk factor for severe COVID-19 patients progressing to critical type, after adjusting for age and gender [adjusted hazard ratio (AHR)=2.38(1.30-4.37), P=0.005] and additionally adjusting for chronic kidney diseases, hypertension, coronary heart disease [AHR=2.63(1.45-4.77), P<0.001], or additionally adjusting for some statistically significant laboratory findings [AHR=2.10(1.13-3.89), P=0.018]. Severe COVID-19 patients with COPD have higher levels of disease severity, proportion of progression to critical illness and mortality rate. Individualized treatment strategies should be adopted to improve the prognosis of severe COVID-19 patients.

Automated diagnosis using artificial intelligence a step forward for preventive dentistry: A systematic review

Background. Early diagnosis and monitoring the evolution of the patients is required to be able to have effective preventive attitudes. An easy and cost-effective way of diagnosis is needed for this purpose. The aim of the study was to evaluate the AI level of use in dentistry diagnosis and the fields of its applicability especially for early diagnosis purposes. A secondary objective was to point out the measured performances for automated AI diagnosis by comparison with standard diagnosis procedures. Material and methods. A comprehensive electronic search was performed in November 2022 through PubMed, Scopus, and Web of Science databases. The following keywords were used to search the databases: (”Artificial Intelligence” OR ”neural network” OR ”Deep learning” OR “Machine learning”) AND (”Dentistry” OR “Dental medicine”) AND (” periodontal disease” OR ”periodontics” OR ”Carious lesions” OR ”oral cancer” OR ”restorative” or “early diagnosis”). The risk of bias (RoB) of the included studies was assessed using PROBAST tool. Results. A total of 334 publications were collected after searching the databases. For 218 remaining publications the title and the abstract were assessed. The reviewers agreed to continue with 69 studies for full text assessment. Because 49 studies had not completely fulfilled the inclusion criteria only 20 publications were included in the final analysis. AI automatic data processing for diagnostic purposes was implemented in the field of dental radiology, oral pathology, restorative dentistry, pedodontics, oncology, endodontics, and periodontics. Conclusion. AI based automatic diagnostic is a powerful and reliable tool that has a great future potential for different fields of dental medicine like periodontal disease, oral cancer, and carious lesions.