Search

Artificial Intelligence (AI) chatbots leveraging Large Language Models (LLMs) are gaining traction in healthcare for their potential to automate patient interactions and aid clinical decision-making. This study examines the reliability of AI chatbots, specifically GPT 4.0, Claude 3 Opus, and Gemini Ultra 1.0, in predicting diseases from patient complaints in the emergency department. The methodology includes few-shot learning techniques to evaluate the chatbots' effectiveness in disease prediction. We also fine-tune the transformer-based model BERT and compare its performance with the AI chatbots. Results suggest that GPT 4.0 achieves high accuracy with increased few-shot data, while Gemini Ultra 1.0 performs well with fewer examples, and Claude 3 Opus maintains consistent performance. BERT's performance, however, is lower than all the chatbots, indicating limitations due to limited labeled data. Despite the chatbots' varying accuracy, none of them are sufficiently reliable for critical medical decision-making, underscoring the need for rigorous validation and human oversight. This study reflects that while AI chatbots have potential in healthcare, they should complement, not replace, human expertise to ensure patient safety. Further refinement and research are needed to improve AI-based healthcare applications' reliability for disease prediction., Comment: 24th IEEE International Conference on Information Reuse and Integration (IEEE IRI 2024), San Jose, CA, USA

The Chief Complaint (CC) is a crucial component of a patient's medical record as it describes the main reason or concern for seeking medical care. It provides critical information for healthcare providers to make informed decisions about patient care. However, documenting CCs can be time-consuming for healthcare providers, especially in busy emergency departments. To address this issue, an autocompletion tool that suggests accurate and well-formatted phrases or sentences for clinical notes can be a valuable resource for triage nurses. In this study, we utilized text generation techniques to develop machine learning models using CC data. In our proposed work, we train a Long Short-Term Memory (LSTM) model and fine-tune three different variants of Biomedical Generative Pretrained Transformers (BioGPT), namely microsoft/biogpt, microsoft/BioGPT-Large, and microsoft/BioGPT-Large-PubMedQA. Additionally, we tune a prompt by incorporating exemplar CC sentences, utilizing the OpenAI API of GPT-4. We evaluate the models' performance based on the perplexity score, modified BERTScore, and cosine similarity score. The results show that BioGPT-Large exhibits superior performance compared to the other models. It consistently achieves a remarkably low perplexity score of 1.65 when generating CC, whereas the baseline LSTM model achieves the best perplexity score of 170. Further, we evaluate and assess the proposed models' performance and the outcome of GPT-4.0. Our study demonstrates that utilizing LLMs such as BioGPT, leads to the development of an effective autocompletion tool for generating CC documentation in healthcare settings., Comment: IEEE BigData 2023 - Sorrento, Italy. 10 Pages, 4 Figures, 5 Tables

The LambdaCDM model is the most commonly accepted framework in modern cosmology. However, the local measurements of the Hubble constant, H0, via the Supernovae Type Ia (SNe Ia) calibrated on Cepheids provide a value which is in significant disagreement, from 4 to 6 sigma, with the value of H0 inferred from the Cosmic Microwave Background (CMB) observed by Planck. This disagreement is the so-called Hubble constant tension. To find out the reason for this discrepancy, we analyze the behaviour of the H0 in the Pantheon sample of SNe Ia through a binning approach: we divide the Pantheon into 3 and 4 bins ordered with redshift (z), and for each of them, we estimate the H0. After the H0 estimation, we fit the H0 values with a decreasing function of z, finding out that H0 undergoes a slow decreasing trend compatible with the evolution scenario in 2.0 sigma. [...] Together with SNe Ia, more astrophysical probes such as quasars (QSO) [...] and Gamma-Ray Bursts (GRBs) [...], are needed to tackle the H0 tension. In the realm of GRB-cosmology, one of the most promising correlations is the fundamental plane relation [...]. In the context of applying this relation as a cosmological tool, we also compute how many GRBs must be gathered to reach the same precision as the SNe Ia. Since we are about two decades away from reaching such precision, we also attempt to find additional correlations for the GRBs associated with SNe Ibc that could be exploited to standardize the class of GRB-SNe Ibc in the future. We find a hint of a correlation between the GRBs' end-of-plateau optical luminosity and the SNe's rest-frame peak time, suggesting that the GRBs with the most luminous optical plateau emission are associated with SNe with the most delayed peaks in their light curves. So far, it is the fundamental plane relation to be the most promising candle for exploring the high-z universe., Comment: 8 pages, 3 figures with 2, 1, and 2 panels respectively. Accepted for publication as a proceeding for The 38th International Cosmic Ray Conference (ICRC2023) through Proceedings of Science (PoS)

We consider the dynamics of strongly localized systems subject to dephasing noise with arbitrary correlation time. Although noise inevitably induces delocalization, transport in the noise-induced delocalized phase is subdiffusive in a parametrically large intermediate-time window. We argue for this intermediate-time subdiffusive regime both analytically and using numerical simulations on single-particle localized systems. Furthermore, we show that normal diffusion is restored in the long-time limit, through processes analogous to variable-range hopping. With numerical simulations based on Lanczos exact diagonalization, we demonstrate that our qualitative conclusions are also valid for interacting systems in the many-body localized phase., Comment: 5 pages, 4 figures + Supplemental Material, v2: additional data on noise-induced dynamics in many-body localized systems