Sheryl Brahnam, Loris Nanni, Shannon McMurtrey, Alessandra Lumini, Rick Brattin, Melinda Slack and Tonya Barrier
Diagnosing pain in neonates is difficult but critical. Although approximately thirty manual pain instruments have been developed for neonatal pain diagnosis, most are complex…
Abstract
Diagnosing pain in neonates is difficult but critical. Although approximately thirty manual pain instruments have been developed for neonatal pain diagnosis, most are complex, multifactorial, and geared toward research. The goals of this work are twofold: 1) to develop a new video dataset for automatic neonatal pain detection called iCOPEvid (infant Classification Of Pain Expressions videos), and 2) to present a classification system that sets a challenging comparison performance on this dataset. The iCOPEvid dataset contains 234 videos of 49 neonates experiencing a set of noxious stimuli, a period of rest, and an acute pain stimulus. From these videos 20 s segments are extracted and grouped into two classes: pain (49) and nopain (185), with the nopain video segments handpicked to produce a highly challenging dataset. An ensemble of twelve global and local descriptors with a Bag-of-Features approach is utilized to improve the performance of some new descriptors based on Gaussian of Local Descriptors (GOLD). The basic classifier used in the ensembles is the Support Vector Machine, and decisions are combined by sum rule. These results are compared with standard methods, some deep learning approaches, and 185 human assessments. Our best machine learning methods are shown to outperform the human judges.
Details
Keywords
Yusra Qamar, Rakesh Kumar Agrawal, Taab Ahmad Samad and Charbel Jose Chiappetta Jabbour
An original systematic review of the academic literature on applications of artificial intelligence (AI) in the human resource management (HRM) domain is carried out to capture…
Abstract
Purpose
An original systematic review of the academic literature on applications of artificial intelligence (AI) in the human resource management (HRM) domain is carried out to capture the current state-of-the-art and prepare an original research agenda for future studies.
Design/methodology/approach
Fifty-nine journal articles are selected based on a holistic search and quality evaluation criteria. By using content analysis and structural concept analysis, this study elucidates the extent and impact of AI application in HRM functions, which is followed by synthesizing a concept map that illustrates how the usage of various AI techniques aids HRM decision-making.
Findings
A comprehensive review of the AI-HRM domain’s existing literature is presented. A concept map is synthesized to present a taxonomical overview of the AI applications in HRM.
Research implications/limitations
An original research agenda comprising relevant research questions is put forward to assist further developments in the AI-HRM domain. An indicative preliminary framework to help transition toward ethical AI is also presented.
Originality/value
This study contributes to the literature through a holistic discussion on the current state of the domain, the extent of AI application in HRM, and its current and perceived future impact on HRM functions. A preliminary ethical framework and an extensive future research agenda are developed to open new research avenues.
Details
Keywords
While ChatGPT is gaining popularity, its potential role in supply chains (SCs) remains unexplored. This study explores the potential applications, benefits and challenges of using…
Abstract
Purpose
While ChatGPT is gaining popularity, its potential role in supply chains (SCs) remains unexplored. This study explores the potential applications, benefits and challenges of using ChatGPT as a tool in SCs.
Design/methodology/approach
The data were gathered through an online survey involving 116 respondents from the academic and industrial sectors who have knowledge of ChatGPT and SC management. These participants were affiliated with the Decision Science Institute (DSI) in the USA and contributed to the published DSI conference proceedings from 2019 to 2022. The survey is structured in three main sections: (1) general information (5 background questions), (2) ChatGPT's potential applications and benefits in SCs (15 pre-determined questions) and (3) potential challenges with using ChatGPT in SCs (5 pre-determined questions). The collected data underwent analysis using IBM SPSS Statistics software.
Findings
ChatGPT can potentially benefit SC operations in 15 areas. Eight potential benefits received more support than the rest, including enhanced process efficiency, cost reduction, providing sustainability reports, better demand forecasting, improved data analysis, streamlined supplier communication, streamlined customer communication, supported promotional activities and enhanced customer satisfaction, but all were supported. Also, the study identified some challenges and hurdles currently impacting the use of ChatGPT in the SC, including that ChatGPT cannot replace experts, it is not an immediate game changer, its uses may lack accuracy, and ChatGPT may take time to reach maturity.
Originality/value
The study is the first to offer empirically grounded evidence of ChatGPT's potential in SCs. The research enhances academic literature by deepening our comprehension of the potential applications of ChatGPT within SCs. Therefore, the study makes an invaluable contribution to the extant literature on ChatGPT in SCs. It can benefit manufacturers, suppliers, logistics providers and other types of businesses through more efficient procurement practices, supplier management, operations and inventory management, logistics practices and customer relationships. Future research may explore how and why ChatGPT is used in SCs.