Monday, June 17, 2024

Beyond p-values: Assessing Statistical Fragility with Robustness and Percent Fragility Indices

While p-values have been the traditional measure of statistical significance in scientific research, they often fail to capture the complexity of statistical evidence. Fragility metrics such as the robustness index and percent fragility index offer a more nuanced approach to evaluating the strength of research findings. By quantifying the impact of small changes in data on statistical significance, these indices provide a valuable complement to p-values. Incorporating fragility metrics into research practices can be seen as a step toward a more responsible form of data-driven decision-making that recognizes the conditional nature of statistical evidence.

Citation: Heston TF. Redefining significance: robustness and percent fragility indices in biomedical research. Stats. 2024;7(2):537-48. 

Wednesday, May 22, 2024

Balancing the Benefits and Risks of Nuclear Medicine: A Safety Imperative

 Nuclear medicine offers remarkable diagnostic and therapeutic capabilities, revolutionizing patient care. However, the use of radioactive materials presents inherent risks that demand unwavering attention to safety protocols. Regulatory authorities, such as the Nuclear Regulatory Commission, establish comprehensive standards for radiopharmaceutical handling, administration, and disposal. Compliance with these regulations requires specialized training for authorized users, meticulous documentation, and regular safety audits. As nuclear medicine continues to evolve, the integration of advanced technologies and the expansion of theranostic agents further emphasize the need for enhanced safety measures. By fostering a culture of education, accountability, and adherence to best practices, the nuclear medicine community can responsibly harness the power of radioactivity to improve patient outcomes while prioritizing safety.

Citation: Heston TF, Tafti D. Nuclear Medicine Safety. [Updated 2024 Mar 14]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 Jan-. Available from:

Thursday, April 25, 2024

The Synergy of AI and Blockchain: Driving the Next Generation of Telemedicine

The convergence of artificial intelligence (AI) in the form of large language models (LLMs) and blockchain technology is poised to drive the next generation of telemedicine. LLMs can rapidly analyze patient records, providing contextualized recommendations and enhancing diagnostic processes. Blockchain technology enables secure, decentralized storage and sharing of medical data, ensuring patient privacy and cross-organizational interoperability. The synergy between these technologies can lead to improved care personalization, automated triage, and secure remote patient monitoring. However, the responsible adoption of AI and blockchain in telemedicine requires addressing challenges such as bias, unclear responsibility, and integration with existing systems while prioritizing patient interests and ethical considerations.

Citation: Heston TF. Perspective Chapter: Integrating Large Language Models and Blockchain in Telemedicine. IntechOpen; 2024. DOI: 10.5772/intechopen.1005063

Wednesday, April 17, 2024

Study Highlights Limitations of ChatGPT-4 in Cardiac Risk Assessment

 Heston and Lewis conducted a study to evaluate the performance of ChatGPT-4 in risk-stratifying patients with atraumatic chest pain. The researchers compared ChatGPT-4's risk scores with established tools like TIMI and HEART scores using simulated patient data. Although the mean scores correlated well, ChatGPT-4 provided different risk scores for identical patient data when presented on separate occasions. This inconsistency suggests that further refinement and customization are necessary before integrating ChatGPT-4 into clinical practice for cardiac risk assessment.

Citation: Heston TF, Lewis LM (2024) ChatGPT provides inconsistent risk-stratification of patients with atraumatic chest pain. PLOS ONE 19(4): e0301854. 

Tuesday, April 2, 2024

Improving Medical Research Communication: The Need for Structured Reporting

In this research article, I suggest adopting structured reporting formats similar to those used in clinical trial abstracts and manuscripts to address the challenges in medical research reporting by online news outlets. Implementing standardized inclusion criteria, such as background information, study methods, main results with statistical analyses, discussion of implications and limitations, and disclosure of conflicts of interest, could enhance the quality and transparency of medical research communication. Collaboration among journalists, news organizations, and medical researchers is crucial to establish and promote best practices, fostering informed public discourse on health topics and ultimately contributing to better health outcomes.

Citation: Heston TF. Critical Gaps in Medical Research Reporting by Online News Media. Cureus. 2024 Apr 2;16(4):e57457. doi: 10.7759/cureus.57457.

Thursday, March 21, 2024

Harnessing the Power of Blockchain and AI for Climate Action

 The fight against climate change requires innovative solutions, and the integration of blockchain technology and large language models (LLMs) holds immense potential. Blockchain ensures the transparency and security of climate data, while LLMs analyze vast datasets to generate actionable insights. This powerful combination enhances data management, improves climate models, and enables evidence-based policymaking. By addressing current challenges and opening up new avenues for collaborative climate action, the synergy between blockchain and AI signifies a transformative shift in the global response to climate change.

Heston TF. A blockchain AI solution to climate change. International Journal of Science and Research Archive. 2024;11(02):450–4. DOI: 

Tuesday, March 5, 2024

The Pitfalls of Statistical Fragility in Survey Research

  Survey research plays a crucial role in understanding complex issues, such as bias and work climate in medicine. However, the validity of the results heavily depends on the robustness of the statistical methods employed. A recent study by Carnes and colleagues serves as a reminder of the pitfalls of statistical fragility. Low response rates, small effect sizes, and multiple significance tests can lead to misleading conclusions. Researchers should strive for higher response rates and larger effect sizes to ensure the reliability of their findings and to drive meaningful change in their fields.

Heston TF. Statistical Fragility in Surveys. Acad Med 2024 Mar 1;99(3):240. doi: 10.1097/ACM.0000000000005584. Epub 2023 Dec 7.