List Of Choices Research Articles

Assessing Large Language Models for Oncology Data Inference From Radiology Reports.

We examined the effectiveness of proprietary and open large language models (LLMs) in detecting disease presence, location, and treatment response in pancreatic cancer from radiology reports. We analyzed 203 deidentified radiology reports, manually annotated for disease status, location, and indeterminate nodules needing follow-up. Using generative pre-trained transformer (GPT)-4, GPT-3.5-turbo, and open models such as Gemma-7B and Llama3-8B, we employed strategies such as ablation and prompt engineering to boost accuracy. Discrepancies between human and model interpretations were reviewed by a secondary oncologist. Among 164 patients with pancreatic tumor, GPT-4 showed the highest accuracy in inferring disease status, achieving a 75.5% correctness (F1-micro). Open models Mistral-7B and Llama3-8B performed comparably, with accuracies of 68.6% and 61.4%, respectively. Mistral-7B excelled in deriving correct inferences from objective findings directly. Most tested models demonstrated proficiency in identifying disease containing anatomic locations from a list of choices, with GPT-4 and Llama3-8B showing near-parity in precision and recall for disease site identification. However, open models struggled with differentiating benign from malignant postsurgical changes, affecting their precision in identifying findings indeterminate for cancer. A secondary review occasionally favored GPT-3.5's interpretations, indicating the variability in human judgment. LLMs, especially GPT-4, are proficient in deriving oncologic insights from radiology reports. Their performance is enhanced by effective summarization strategies, demonstrating their potential in clinical support and health care analytics. This study also underscores the possibility of zero-shot open model utility in environments where proprietary models are restricted. Finally, by providing a set of annotated radiology reports, this paper presents a valuable data set for further LLM research in oncology.

Can Climate Shocks Make Vulnerable Subjects More Willing to Take Risks?

While economists in the past tended to assume that individual preferences, including risk preferences, are stable over time, a recent literature has developed and indicates that risk preferences respond to shocks, with mixed evidence on the direction of the responses. This paper utilizes a natural experiment with covariate (drought) and idiosyncratic shocks in combination with an independent field risk experiment. The risk experiment uses a Certainty Equivalent-Multiple Choice List approach and is played 1–2 years after the subjects were (to a varying degree) exposed to a covariate drought shock or idiosyncratic shocks for a sample of resource-poor young adults living in a risky semi-arid rural environment in Sub-Saharan Africa. The experimental approach facilitates a comprehensive assessment of shock effects on experimental risk premiums for risky prospects with varying probabilities of good and bad outcomes. The experiment also facilitates the estimation of the utility curvature in an Expected Utility (EU) model and, alternatively, separate estimation of probability weighting and utility curvature in three different Rank Dependent Utility models with a two-parameter Prelec probability weighting function. Our study is the first to comprehensively test the theoretical predictions of Gollier and Pratt (Econom J Econom Soc 64:1109–1123, 1996) versus Quiggin (Econ Theor 22(3):607–611, 2003). Gollier and Pratt (1996) build on EU theory and state that an increase in background risk will make subjects more risk averse while Quiggin (2003) states that an increase in background risk can enhance risk-taking in certain types of non-EU models. We find strong evidence that such non-EU preferences dominate in our sample.

Assessing the use of HL7 FHIR for implementing the FAIR guiding principles: a case study of the MIMIC-IV Emergency Department module.

To provide a real-world example on how and to what extent Health Level Seven Fast Healthcare Interoperability Resources (FHIR) implements the Findable, Accessible, Interoperable, and Reusable (FAIR) guiding principles for scientific data. Additionally, presents a list of FAIR implementation choices for supporting future FAIR implementations that use FHIR. A case study was conducted on the Medical Information Mart for Intensive Care-IV Emergency Department (MIMIC-ED) dataset, a deidentified clinical dataset converted into FHIR. The FAIRness of this dataset was assessed using a set of common FAIR assessment indicators. The FHIR distribution of MIMIC-ED, comprising an implementation guide and demo data, was more FAIR compared to the non-FHIR distribution. The FAIRness score increased from 60 to 82 out of 95 points, a relative improvement of 37%. The most notable improvements were observed in interoperability, with a score increase from 5 to 19 out of 19 points, and reusability, with a score increase from 8 to 14 out of 24 points. A total of 14 FAIR implementation choices were identified. Our work examined how and to what extent the FHIR standard contributes to FAIR data. Challenges arose from interpreting the FAIR assessment indicators. This study stands out for providing a real-world example of a dataset that was made more FAIR using FHIR. To the best of our knowledge, this is the first study that formally assessed the conformance of a FHIR dataset to the FAIR principles. FHIR improved the accessibility, interoperability, and reusability of MIMIC-ED. Future research should focus on implementing FHIR in research data infrastructures.

Neural Prompt Search.

The size of vision models has grown exponentially over the last few years, especially after the emergence of Vision Transformer. This has motivated the development of parameter-efficient tuning methods, such as learning adapter layers or visual prompt tokens, which allow a tiny portion of model parameters to be trained whereas the vast majority obtained from pre-training are frozen. However, designing a proper tuning method is non-trivial: one might need to try out a lengthy list of design choices, not to mention that each downstream dataset often requires custom designs. In this paper, we view the existing parameter-efficient tuning methods as "prompt modules" and propose Neural prOmpt seArcH (NOAH), a novel approach that learns, for large vision models, the optimal design of prompt modules through a neural architecture search algorithm, specifically for each downstream dataset. By conducting extensive experiments on over 20 vision datasets, we demonstrate that NOAH (i) is superior to individual prompt modules, (ii) has good few-shot learning ability, and (iii) is domain-generalizable. The code and models are available at https://github.com/ZhangYuanhan-AI/NOAH.

On a Vernam cipher scheme based on elliptic curve L-functions

In this paper, we present a new Vernam-like cryptosystem based on the L-functions attached to elliptic curves. Actually, we show that for infinitely many elliptic curves, the coefficients of the L-series can be considered as a secure cryptographic cipher key when combined with linear and classical pseudo-random number generators. Hence, fast and widely known classical generators, like linear congruential generators, should be reintroduced into our list of cryptographic design choices.

Design and requirements of a hydrogen component reliability database (HyCReD)

Hydrogen technologies are expected to play a key role in the decarbonization of several sectors including energy storage and transportation. Rigorous investigation and quantification of the risk and reliability issues associated with hydrogen technologies will be critical to ensuring both their wider adoption and safe, economical operations. Quantitative risk assessment (QRA) is an important tool that has been used to enable the safe deployment of many engineering systems, including hydrogen fueling stations and hydrogen storage systems. However, QRA studies require reliability data which is currently lacking for expanding applications of hydrogen systems. To address this gap, we present a new structure for a hydrogen component reliability database (HyCReD) that can be used to generate reliability data to be used in QRA, reliability, safety studies, maintenance planning, and more. Building on our previous work examining four major hydrogen safety data collection tools (West et al., 2022) [1], our approach in this work was to consult scientific literature on reliability data collection as well as a number of existing reliability engineering databases in the oil & gas, chemical processing, and nuclear power plant sectors. The evaluation of these databases led to identifying best practices to be implemented in a data collection framework for a hydrogen component reliability database. Based on these best practices, a set of 24 requirements for the proposed database are presented, covering its characteristics and the types of data to be collected. We define the structure of the HyCReD database and 25 data elements to be collected, spanning system description, failure, shutdown, or near-miss events, and maintenance events. The data elements are then defined according to international standards used in the safety and reliability practice and potential choice lists are provided for each field. Since this database is being piloted for hydrogen fueling stations, a generic station component hierarchy developed by West (2021) [2] is used to standardize system data. Finally, we demonstrate populating the database with information extracted from five narrative reports on hydrogen fueling station incidents.

Against All Odds

Against All Odds

An Artificial Intelligence Tool for Clinical Decision Support and Protocol Selection for Brain MRI.

Protocolling, the process of determining the most appropriate acquisition parameters for an imaging study, is time-consuming and produces variable results depending on the performing physician. The purpose of this study was to assess the potential of an artificial intelligence-based semiautomated tool in reducing the workload and decreasing unwarranted variation in the protocolling process. We collected 19,721 MR imaging brain examinations at a large academic medical center. Criterion standard labels were created using physician consensus. A model based on the Long Short-Term Memory network was trained to predict the most appropriate protocol for any imaging request. The model was modified into a clinical decision support tool in which high-confidence predictions, determined by the values the model assigns to each possible choice, produced the best protocol automatically and low confidence predictions provided a shortened list of protocol choices for review. The model achieved 90.5% accuracy in predicting the criterion standard labels and demonstrated higher agreement than the original protocol assignments, which achieved 85.9% accuracy (κ = 0.84 versus 0.72, P value < .001). As a clinical decision support tool, the model automatically assigned 70% of protocols with 97.3% accuracy and, for the remaining 30% of examinations, achieved 94.7% accuracy when providing the top 2 protocols. Our model achieved high accuracy on a standard based on physician consensus. It showed promise as a clinical decision support tool to reduce the workload by automating the protocolling of a sizeable portion of examinations while maintaining high accuracy for the remaining examinations.

A novel value-based multi-criteria decision making approach to evaluate new technology adoption in SMEs.

The modern competition is moving quickly toward incorporating cutting-edge technological improvements to support Small and Medium Enterprises (SMEs) in enhancing their business models. Making decisions regarding implementing new technologies in SMEs is a challenging process driven by the continuous new advances in the technology industry. Important operational process decisions, such as organizational investment costs, technology acquisition, maintenance, customer experience, employee training for the proper use of each technology, reliability requirements, and security needs, must be made very carefully for such adoptions. In this research, a novel multi-criteria decision-making process model is presented to help SME decision-makers choose the optimal technology from a list of desirable choices. The proposed approach makes use of a weighted multi-criteria decision model to rank parameters by combining many criteria that are crucial in choosing the best-suited technology. A literature review and expert opinion are utilized to pinpoint key decision-makers for the adoption of technological innovation. Specialists had reviewed the proposal in the field, and early findings suggest that it might be helpful to SME decision-makers in promoting customer value and firm performance.

Automated Clinicopathologic Correlation Integrating Clinical Findings, Suspicions, Hematopathology Tests and Previous Records

Abstract Introduction/Objective Conducting clinicopathologic correlation (CPC) to provide comprehensive diagnosis, prognosis/risk assessment, follow up and/or recommendations is essential in Hematopathology. The clinical findings/suspicions and medical history contribute in building a personalized diagnosis and prognosis, while comparing with the previous records is required for specialized follow-ups and progression assessment. HIPAA violation liability and changeable medical guidelines limit the "freelance outsourcing" of Hematopathology assistance/second opinion. The proposed electronic Hematopathologist assistant could play this role perfectly, because automation in Hematopathology is easier when compared with other medical specialties. This is because only visual sense is needed for Hematopathology review/examination while automating more than one sense is needed in medicine and surgery. Methods/Case Report Scalable, customizable and fully controlled system is developed harnessing the state-of-the-art technologies. The user selects specific answers in smart hierarchical choice lists about specimen types, medical history, clinical findings and suspicions. Then, the available current and previous results of the performed diagnostic and prognostic test panels are needed to be entered and dated. These panels include quantitative and qualitative parameters of the complete blood count (CBC), smear review, flow cytometry, immunohistochemistry, cytogenetics and/or molecular studies. Zero harm to the patients privacy is achieved, there is no personal information entered or needed. Results (if a Case Study enter NA) Based on the entered results, detailed interpretation for each test panel is created to explain how the generated comprehensive CPC is built. Furthermore, the CPC is concluded in one line, this is an informative and summative title for the case. The system is also used as a triage advisor, it suggests the useful parameters to be triaged considering the choosing wisely recommendations and providing the potential percentage of resulting positive/abnormal in similar specific cases. All possible generated test panel interpretations, CPCs and case titles are reviewed and validated. Conclusion This system generates preliminary versions for the Hematopathologist to review, leading to productivity increase with no medical liability or feeling fatigued and stressed. This also helps to increase the profitability and cover the shortage.

On Comprehension of Genetic Programming Solutions: A Controlled Experiment on Semantic Inference

Applied to the problem of automatic program generation, Genetic Programming often produces code bloat, or unexpected solutions that are, according to common belief, difficult to comprehend. To study the comprehensibility of the code produced by Genetic Programming, attribute grammars obtained by Genetic Programming-based semantic inference were compared to manually written ones. According to the established procedure, the research was carried out as a controlled classroom experiment that involved two groups of students from two universities, and consisted of a background questionnaire, two tests and a feedback questionnaire after each test. The tasks included in the tests required the identification of various properties of attributes and grammars, the identification of the correct attribute grammar from a list of choices, or correcting a semantic rule in an attribute grammar. It was established that solutions automatically generated by Genetic Programming in the field of semantic inference, in this study attribute grammars, are indeed significantly harder to comprehend than manually written ones. This finding holds, regardless of whether comprehension correctness, i.e., how many attribute grammars were correctly comprehended, or comprehension efficiency is considered, i.e., how quickly attribute grammars were correctly comprehended.

ATTITUDE TOWARDS PSYCHIATRY AND PSYCHIATRISTS AMONG MEDICAL STUDENTS IN TERTIARY HEALTH CARE HOSPITAL

Background. Psychiatry is a medical subject; many medical students ignore its importance in medicine. Regarding mental illness, many medical professionals and students are less aware as mental illness is very important in affecting the quality of care the patient receives and in choosing psychiatry as a career. This study specializes in psychiatry based on many factors, however undergraduate students’ attitude towards career choice is the most important. Objective. This study aims to explore attitude towards psychiatry subject among undergraduate medical students. Methods. This is a Descriptive cross-sectional study done at Vikhe Patil Medical College, Ahmednagar. The total of 295 volunteer students from all phases of MBBS (Bachelor of Medicine and Bachelor of Surgery) were enrolled by purposive sampling, a 30-item self-administered questionnaire, ATP-30 (Attitude towards Psychiatry) was used to measure the attitude of medical students. Data collected was analysed using SPSS version 24. The results were reported as percentage and frequencies. Results. In the study, 47.4% of the respondents were males and 52.6% were females. Out of total study participants 52.3% of our study respondents strongly agreed that psychiatry was a respected branch of medicine; 56.7% strongly agreed that psychiatry should be an important part of medical curriculum; 19.4% of study respondents considered it in the list of the career choice while the majority did not think that it should be in their career list options. Conclusions. In the study, the students from the later phases of MBBS showed positive attitude than that of the initial phases. Many students found it as respected subject and essential field of medicine for their choice of profession. However, shortcoming knowledge and awareness was observed in the responses pertaining to present available treatment and recent trends in the field.

The Pediatric Pharmacy Association and Members Threaded Through a Career and the Impact on Pediatric Drug Development.

The Pediatric Pharmacy Association and Members Threaded Through a Career and the Impact on Pediatric Drug Development.

Reevaluating Gendered Language in Surgery.

Reevaluating Gendered Language in Surgery.

Studi Komparatif Fungsionalitas, Performance dan User experience pada Website Geopark

Indonesia has great natural resources. UNESCO recognized at least six geoparks, and Indonesia will continue to develop more geoparks in the future. However, currently, Indonesia's geoparks are not widely known. One of the reasons is that the existing websites do not pay attention to performance guidelines, information, economics, and search engine optimization even though the website is a vital marketing medium so that geoparks can be promoted widely and effectively. A comprehensive study of the website contents is required to build a good geopark website. In addition, it is also necessary to study user experience when visiting web pages. This study aims to find the standard website content on the geopark website and compare existing website performance and user experience. This study compares three geopark websites. The three websites include the Ciletuh-Palabuhanratu, Hongkong geopark, and Cliff of Fundy Geosite website. Three indicators are used: website performance, user experience, and what information should be on the geopark website. These three indicators are essential as an attraction for website visitors so they can enjoy the content of the website and they can stay longer so that the visitors can turn into potential geopark tourists. Good website performance is supported by consistent ease of providing information and speed in providing information to website visitors. In comparison, the user experience visiting the website relates to the interactive visual design, organization, and navigation structure of the website framework and the completeness of the content. We use performance and user experience metrics as evaluation tools for the three selected websites. As a result, the general availability of information is as follows: information about geoparks, UNESCO, list of tourist attractions, activities, itinerary, tourist accommodation, education, contacts, language choices, news articles, pamphlets, maps, and virtual tours. In terms of performance, the three websites have slow speeds, but in cumulative layout shifts, they have good scores.

Perceived awareness about mental illness among journalism students and medical students

Aim: The study was planned to assess the perceived level of awareness about mental illness among the Journalism students and the Medical students. Materials and methods: A total of 210 students were surveyed, 70 in each group to know their perception of awareness about mental illness. The questionnaire consisted of questions regarding contact with psychiatrist during their lifetime for self, family and friends, about exposure to psychiatry/psychology as one of the subjects during their course. They were also asked about the perceived cause of mental illness and source of information regarding mental illness from a list of choices. Results: Only 9 out of 210 respondents stated that they have a family history of mental illness, however, almost 31 respondents stated that they have come in contact with a psychiatrist in their lifetime with regard to consultation for self, family or friend. Among the JPG students, 59.5% felt that the portrayal in the movies was in negative light. Regarding the portrayal of mental illness in media, majority of journalism students believed that it was portrayed positively (JUG-97.2% and JPG-81.1%). In contrast, only 62.5% of the medical students believed that mental illness is portrayed positively in media.

Technology sourcing and the dark side of open innovation: Evidence from the biopharmaceutical sector

Firms in high-technology industries face a complex set of challenges to innovate successfully and continuously to gain a sustainable competitive advantage. Innovative new R&D project sourcing choices top this list of difficult and nuanced choices that firms must make in pursuit of innovation. Our research makes a significant and novel contribution to this interesting discourse and critically examines the sourcing decision (i.e., open vs. closed innovation) in the context of new product development in the biopharmaceutical sector. Specifically, we apply a project-level typology along the dimensions of new R&D project source and project familiarity. Drawing from transaction cost economics (TCE) and the knowledge-based view (KBV) theories, we empirically test our theoretically developed hypotheses on a dataset of 2,971 biopharmaceutical R&D projects spanning from 1985 to 2016. Results from our robust analysis show that both new R&D project source and project familiarity have significant direct effects on focal project performance outcomes (defined in this paper as likelihood of termination). We also determine that focal project familiarity has a significant moderating effect on the relationship between project source and performance outcome.

Reference Dependence in Intertemporal Preference

Dynamic structures in intertemporal choices offer multiple channels through which reference dependence can occur. We examine intertemporal reference dependence when using choice lists to elicit time preference over outcome sequences. Reference point effects can arise endogenously within an outcome sequence and exogenously from the experimental environment of the choice list. We show that estimated discount factors in the choice list are biased by both kinds of reference points and propose a model to jointly account for both. Our experimental design with outcome sequences also enables us to jointly estimate the discount factor and utility curvature. We further discuss the implications for recent developments in the measurement of intertemporal preference. This paper was accepted by Manel Baucells, behavioral economics and decision analysis. Funding: S. Zhong was supported by the National University of Singapore [Singapore Ministry of Education Academic Research Fund Tier 1 Grant R-122-000-325-115]. Supplemental Material: The data files and online appendices are available at https://doi.org/10.1287/mnsc.2022.4348 .

Design of Interactive Protocols that Help Students to Prepare for Laboratory Work

Laboratory education makes up an extensive part of natural science education at universities. To support meaningful learning during laboratory work, students should prepare by carefully studying the protocols. The aim of this study was to design and evaluate a preparative learning material that focuses on protocol steps with the design goal to help students to (i) understand the “why” (reason or theory) of protocol steps, (ii) make practical choices, and (iii) do troubleshooting. To reach this goal, protocol steps have been enriched with theoretical, practical, troubleshooting, and/or calculation questions to form interactive protocols. This design process resulted in a list of design choices (i.e., when to include a question and how to design a question) and a showcase of questions in the interactive protocols. These interactive protocols were implemented and evaluated in a real educational setting. From the evaluation results, it was concluded that the interactive protocols were successful in preparing students for the laboratory work. After the laboratory work, students reported a more diverse and slightly less positive contribution of the interactive protocols to their understanding of the protocol steps and their ability to make practical choices. A significant difference was found in the perceived usefulness of the troubleshooting questions before versus after the laboratory work, which is suggested as a topic for further investigation.

Factor Construction Zoo: Are Factor Exposures Created Equal?

The answer is no. Factor investing provides investors with a low-cost avenue to participate in stock selection. Investors earn excess returns for taking on factor risk, which is often measured by factor exposure. However, the relation between the return and factor exposure is nonlinear. Large-scale simulation shows that similar target factor exposures can be engineered using various portfolio construction methodologies, but the resulting portfolios exhibit significant dispersion in expected returns and co-movement with the market and across factor funds. The dispersion increases with target factor exposures. As such, some factor exposures are more efficient than others. This article further studies a comprehensive list of portfolio construction choices for value, momentum, and quality funds; discusses the trade-off at work; and provides a framework for the assessment of factor exposure efficiency. It is important to account for nonlinearity in constructing or evaluating factor funds.