Project description:Resume screening assisted by decision support systems that incorporate artificial intelligence is currently undergoing a strong development in many organizations, raising technical, managerial, legal, and ethical issues. The purpose of the present paper is to better understand the reactions of recruiters when they are offered algorithm-based recommendations during resume screening. Two polarized attitudes have been identified in the literature on users' reactions to algorithm-based recommendations: algorithm aversion, which reflects a general distrust and preference for human recommendations; and automation bias, which corresponds to an overconfidence in the decisions or recommendations made by algorithmic decision support systems (ADSS). Drawing on results obtained in the field of automated decision support areas, we make the general hypothesis that recruiters trust human experts more than ADSS, because they distrust algorithms for subjective decisions such as recruitment. An experiment on resume screening was conducted on a sample of professionals (N = 694) involved in the screening of job applications. They were asked to study a job offer, then evaluate two fictitious resumes in a 2 × 2 factorial design with manipulation of the type of recommendation (no recommendation/algorithmic recommendation/human expert recommendation) and of the consistency of the recommendations (consistent vs. inconsistent recommendation). Our results support the general hypothesis of preference for human recommendations: recruiters exhibit a higher level of trust toward human expert recommendations compared with algorithmic recommendations. However, we also found that recommendation's consistence has a differential and unexpected impact on decisions: in the presence of an inconsistent algorithmic recommendation, recruiters favored the unsuitable over the suitable resume. Our results also show that specific personality traits (extraversion, neuroticism, and self-confidence) are associated with a differential use of algorithmic recommendations. Implications for research and HR policies are finally discussed.
Project description:To address the issues of data sparsity, scalability, and cold start in the traditional teaching resource recommendation process, this paper presents an enhanced collaborative filtering (CF) recommendation algorithm incorporating a time decay (TD) function. By aligning with the human memory forgetting curve, the TD function is employed as a weighting factor, enabling the calculation of similarity and user preferences constrained by the TD, thus amplifying the weight of user interest over a short period and achieving the integration of short-term and long-term interests. The results indicate that the RMSE of the proposed combined recommendation algorithm (TD-CF) is only 8.95 when the number of recommendations reaches 100, which is significantly lower than the comparison model, which exhibits higher accuracy across different recommended items, effectively utilizing music teaching resources and user characteristics to deliver more precise recommendations.
Project description:BackgroundIn medical imaging courses, due to the complexity of anatomical relationships, limited number of practical course hours and instructors, how to improve the teaching quality of practical skills and self-directed learning ability has always been a challenge for higher medical education. Artificial intelligence-assisted diagnostic (AISD) software based on volume data reconstruction (VDR) technique is gradually entering radiology. It converts two-dimensional images into three-dimensional images, and AI can assist in image diagnosis. However, the application of artificial intelligence in medical education is still in its early stages. The purpose of this study is to explore the application value of AISD software based on VDR technique in medical imaging practical teaching, and to provide a basis for improving medical imaging practical teaching.MethodsTotally 41 students majoring in clinical medicine in 2017 were enrolled as the experiment group. AISD software based on VDR was used in practical teaching of medical imaging to display 3D images and mark lesions with AISD. Then annotations were provided and diagnostic suggestions were given. Also 43 students majoring in clinical medicine from 2016 were chosen as the control group, who were taught with the conventional film and multimedia teaching methods. The exam results and evaluation scales were compared statistically between groups.ResultsThe total skill scores of the test group were significantly higher compared with the control group (84.51 ± 3.81 vs. 80.67 ± 5.43). The scores of computed tomography (CT) diagnosis (49.93 ± 3.59 vs. 46.60 ± 4.89) and magnetic resonance (MR) diagnosis (17.41 ± 1.00 vs. 16.93 ± 1.14) of the experiment group were both significantly higher. The scores of academic self-efficacy (82.17 ± 4.67) and self-directed learning ability (235.56 ± 13.50) of the group were significantly higher compared with the control group (78.93 ± 6.29, 226.35 ± 13.90).ConclusionsApplying AISD software based on VDR to medical imaging practice teaching can enable students to timely obtain AI annotated lesion information and 3D images, which may help improve their image reading skills and enhance their academic self-efficacy and self-directed learning abilities.
Project description:Artificial intelligence algorithms to classify melanoma are dependent on their training data, which limits generalizability. The objective of this study was to compare the performance of an artificial intelligence model trained on a standard adult-predominant dermoscopic dataset before and after the addition of additional pediatric training images. The performances were compared using held-out adult and pediatric test sets of images. We trained two models: one (model A) on an adult-predominant dataset (37,662 images from the International Skin Imaging Collaboration) and the other (model A+P) on an additional 1,536 pediatric images. We compared performance between the two models on adult and pediatric held-out test images separately using the area under the receiver operating characteristic curve. We then used Gradient-weighted Class Activation Maps and background skin masking to understand the contributions of the lesion versus background skin to algorithm decision making. Adding images from a pediatric population with different epidemiological and visual patterns to current reference standard datasets improved algorithm performance on pediatric images without diminishing performance on adult images. This suggests a way that dermatologic artificial intelligence models can be made more generalizable. The presence of background skin was important to the pediatric-specific improvement seen between models. Our study highlights the importance of carefully curated and labeled data from diverse inputs to improve the generalizability of AI models for dermatology, in this case applied to dermoscopic images of adult and pediatric lesions to improve melanoma detection.
Project description:IntroductionArtificial intelligence and neuroimaging enable accurate dementia prediction, but 'black box' models can be difficult to trust. Explainable artificial intelligence (XAI) describes techniques to understand model behaviour and the influence of features, however deciding which method is most appropriate is non-trivial. Vision transformers (ViT) have also gained popularity, providing a self-explainable, alternative to traditional convolutional neural networks (CNN).MethodsWe used T1-weighted MRI to train models on two tasks: Alzheimer's disease (AD) classification (diagnosis) and predicting conversion from mild-cognitive impairment (MCI) to AD (prognosis). We compared ten XAI methods across CNN and ViT architectures.ResultsModels achieved balanced accuracies of 81% and 67% for diagnosis and prognosis. XAI outputs highlighted brain regions relevant to AD and contained useful information for MCI prognosis.DiscussionXAI can be used to verify that models are utilising relevant features and to generate valuable measures for further analysis.
Project description:Previous studies have shown that the manufacturer's default preoperative plans for total knee arthroplasty with patient-specific guides require frequent, time-consuming changes by the surgeon. Currently, no research has been done on predicting preoperative plans for orthopedic surgery using machine learning. Therefore, this study aims to evaluate whether artificial intelligence (AI) driven planning tools can create surgeon and patient-specific preoperative plans that require fewer changes by the surgeon. A dataset of 5409 preoperative plans, including the manufacturer's default and the plans corrected by 39 surgeons, was collected. Features were extracted from the preoperative plans that describe the implant sizes, position, and orientation in a surgeon- and patient-specific manner. Based on these features, non-linear regression models were employed to predict the surgeon's corrected preoperative plan. The average number of corrections a surgeon has to make to the preoperative plan generated using AI was reduced by 39.7% compared to the manufacturer's default plan. The femoral and tibial implant size in the manufacturer's plan was correct in 68.4% and 73.1% of the cases, respectively, while the AI-based plan was correct in 82.2% and 85.0% of the cases, respectively, compared to the surgeon approved plan. Our method successfully demonstrated the use of machine learning to create preoperative plans in a surgeon- and patient-specific manner for total knee arthroplasty.
Project description:This study aims to evaluate the practical application value of the teaching method under the guidance of educational psychology and artificial intelligence (AI) design, taking the deep learning theory as the basis of teaching design. The research objects of this study involve all the teachers, students, and students' parents of Ningbo Middle School. The questionnaires are developed to survey the changes in the performance of students before and after the implementation of the teaching design and the satisfaction of all teachers, students, and parents to different teaching methods by comparing the two results and the satisfaction ratings. All objects in this study volunteer to participate in the questionnaire survey. The results suggest the following: (1) the effective return rates of the questionnaires to teachers, students, and parents are 97, 99, and 95%, respectively, before implementation; whereas those after implementation are 98, 99, and 99%, respectively. Comparison of the two return results suggests that there was no significant difference statistically (P > 0.05). (2) Proportion of scoring results before and after implementation is given as follows: the proportions of levels A, B, C, and D are 35, 40, 15, and 10% before implementation, respectively; while those after implementation are 47, 36, 12, and 5%, respectively. After the implementation, the proportion of level A is obviously higher than that before the implementation, and the proportions of other levels decreased in contrast to those before the implementation, showing statistically obvious differences (P < 0.05). (3) The change in the performance of each subject after 1 year implementation is significantly higher than that before the implementation, and the change in the average performance of each subject shows an upward trend. In summary, (1) the comparison on the effective return rate of the satisfaction survey questionnaire proves the feasibility of its scoring results. (2) The comparison of the survey scoring results shows that people are more satisfied with the new educational design teaching method. (3) The comparison of the change in the performance of each subject before and after the implementation indirectly reflects the drawbacks of partial subject education, indicating that the school should pay the same equal attention to every subject. (4) Due to various objective and subjective factors, the results of this study may be different from the actual situation slightly, and its accuracy has to be further explored in the future.
Project description:ImportanceEarly breast cancer detection is associated with lower morbidity and mortality.ObjectiveTo examine whether a commercial artificial intelligence (AI) algorithm for breast cancer detection could estimate the development of future cancer.Design, setting, and participantsThis retrospective cohort study of 116 495 women aged 50 to 69 years with no prior history of breast cancer before they underwent at least 3 consecutive biennial screening examinations used scores from an AI algorithm (INSIGHT MMG, version 1.1.7.2; Lunit Inc; used September 28, 2022, to April 5, 2023) for breast cancer detection and screening data from multiple, consecutive rounds of mammography performed from September 13, 2004, to December 21, 2018, at 9 breast centers in Norway. The statistical analyses were performed from September 2023 to August 2024.ExposureArtificial intelligence algorithm score indicating suspicion for the presence of breast cancer. The algorithm provided a continuous cancer detection score for each examination ranging from 0 to 100, with increasing values indicating a higher likelihood of cancer being present on the current mammogram.Main outcomes and measuresMaximum AI algorithm score for cancer detection and absolute difference in score among breasts of women developing screening-detected cancer, women with interval cancer, and women who screened negative.ResultsThe mean (SD) age at the first study round was 58.5 (4.5) years for 1265 women with screening-detected cancer in the third round, 57.4 (4.6) years for 342 women with interval cancer after 3 negative screening rounds, and 56.4 (4.9) years for 116 495 women without breast cancer all 3 screening rounds. The mean (SD) absolute differences in AI scores among breasts of women developing screening-detected cancer were 21.3 (28.1) at the first study round, 30.7 (32.5) at the second study round, and 79.0 (28.9) at the third study round. The mean (SD) differences prior to interval cancer were 19.7 (27.0) at the first study round, 21.0 (27.7) at the second study round, and 34.0 (33.6) at the third study round. The mean (SD) differences among women who did not develop breast cancer were 9.9 (17.5) at the first study round, 9.6 (17.4) at the second study round, and 9.3 (17.3) at the third study round. Areas under the receiver operating characteristic curve for the absolute difference were 0.63 (95% CI, 0.61-0.65) at the first study round, 0.72 (95% CI, 0.71-0.74) at the second study round, and 0.96 (95% CI, 0.95-0.96) at the third study round for screening-detected cancer and 0.64 (95% CI, 0.61-0.67) at the first study round, 0.65 (95% CI, 0.62-0.68) at the second study round, and 0.77 (95% CI, 0.74-0.79) at the third study round for interval cancers.Conclusions and relevanceIn this retrospective cohort study of women undergoing screening mammography, mean absolute AI scores were higher for breasts developing vs not developing cancer 4 to 6 years before their eventual detection. These findings suggest that commercial AI algorithms developed for breast cancer detection may identify women at high risk of a future breast cancer, offering a pathway for personalized screening approaches that can lead to earlier cancer diagnosis.
Project description:AimsThe increasing use of insertable cardiac monitors (ICM) produces a high rate of false positive (FP) diagnoses. Their verification results in a high workload for caregivers. We evaluated the performance of an artificial intelligence (AI)-based ILR-ECG Analyzer™ (ILR-ECG-A). This machine-learning algorithm reclassifies ICM-transmitted events to minimize the rate of FP diagnoses, while preserving device sensitivity.Methods and resultsWe selected 546 recipients of ICM followed by the Implicity™ monitoring platform. To avoid clusterization, a single episode per ICM abnormal diagnosis (e.g. asystole, bradycardia, atrial tachycardia (AT)/atrial fibrillation (AF), ventricular tachycardia, artefact) was selected per patient, and analyzed by the ILR-ECG-A, applying the same diagnoses as the ICM. All episodes were reviewed by an adjudication committee (AC) and the results were compared. Among 879 episodes classified as abnormal by the ICM, 80 (9.1%) were adjudicated as 'Artefacts', 283 (32.2%) as FP, and 516 (58.7%) as 'abnormal' by the AC. The algorithm reclassified 215 of the 283 FP as normal (76.0%), and confirmed 509 of the 516 episodes as abnormal (98.6%). Seven undiagnosed false negatives were adjudicated as AT or non-specific abnormality. The overall diagnostic specificity was 76.0% and the sensitivity was 98.6%.ConclusionThe new AI-based ILR-ECG-A lowered the rate of FP ICM diagnoses significantly while retaining a > 98% sensitivity. This will likely alleviate considerably the clinical burden represented by the review of ICM events.