Journey through diagnosis.

This reflective poem explores the profound human impact of receiving a serious medical diagnosis. The speaker grapples with the emotional upheaval of this sudden severing from one's presumed healthy future. There are attempts to cling to denial or bargain for a different outcome. But the truth of the diagnosis persists, sending ripples of change throughout the patient's life. Dreams slip away and plans evaporate in the crucible of illness. After a struggle, the mind makes peace and courageously leans into the difficulties ahead. The poem celebrates the human capacity to accept vulnerability, find gifts within trials, and walk the remaining road with wisdom. It reflects on how a diagnosis can heighten awareness that life is fleeting and precious. The accompanying digital artwork was generated using OpenAI's DALL·E 3 and modified using Adobe Firefly. It is a stark, black canvas, which can be seen as a metaphor for the profound and contemplative journey described patients go through. It symbolizes the inner darkness and uncertainty faced when confronting life-altering diagnoses, echoing the feelings of isolation, the search for meaning, and the gradual acceptance of a new reality as one navigates through the trials of illness.

Large Language Models in Ophthalmology: Potential and Pitfalls.

Large language models (LLMs) show great promise in assisting clinicians in general, and ophthalmology in particular, through knowledge synthesis, decision support, accelerating research, enhancing education, and improving patient interactions. Specifically, LLMs can rapidly summarize the latest literature to keep clinicians up-to-date. They can also analyze patient data to highlight crucial insights and recommend appropriate tests or referrals. LLMs can automate tedious research tasks like data cleaning and literature reviews. As AI tutors, LLMs can fill knowledge gaps and assess competency in trainees. As chatbots, they can provide empathetic, personalized responses to patient inquiries and improve satisfaction. The visual capabilities of LLMs like GPT-4 allow assisting the visually impaired by describing environments. However, there are significant ethical, technical, and legal challenges around the use of LLMs that should be addressed regarding privacy, fairness, robustness, attribution, and regulation. Ongoing oversight and refinement of models is critical to realize benefits while minimizing risks and upholding responsible AI principles. If carefully implemented, LLMs hold immense potential to push the boundaries of care, discovery, and quality of life for ophthalmology patients.

A scientometric analysis of fairness in health AI literature.

Artificial intelligence (AI) and machine learning are central components of today's medical environment. The fairness of AI, i.e. the ability of AI to be free from bias, has repeatedly come into question. This study investigates the diversity of members of academia whose scholarship poses questions about the fairness of AI. The articles that combine the topics of fairness, artificial intelligence, and medicine were selected from Pubmed, Google Scholar, and Embase using keywords. Eligibility and data extraction from the articles were done manually and cross-checked by another author for accuracy. Articles were selected for further analysis, cleaned, and organized in Microsoft Excel; spatial diagrams were generated using Public Tableau. Additional graphs were generated using Matplotlib and Seaborn. Linear and logistic regressions were conducted using Python to measure the relationship between funding status, number of citations, and the gender demographics of the authorship team. We identified 375 eligible publications, including research and review articles concerning AI and fairness in healthcare. Analysis of the bibliographic data revealed that there is an overrepresentation of authors that are white, male, and are from high-income countries, especially in the roles of first and last author. Additionally, analysis showed that papers whose authors are based in higher-income countries were more likely to be cited more often and published in higher impact journals. These findings highlight the lack of diversity among the authors in the AI fairness community whose work gains the largest readership, potentially compromising the very impartiality that the AI fairness community is working towards.

Familial retinoblastoma: variations in clinical presentation and management based on paternal versus maternal inheritance.

BACKGROUND: Several studies have demonstrated the effect of parent-of-origin on retinoblastoma penetrance. The purpose of the current study was to assess differences in clinical presentation of paternally versus maternally inherited retinoblastoma. METHODS: The clinical records of all children with familial retinoblastoma treated on a tertiary Ocular Oncology Service between December 1975 and May 2020 were reviewed retrospectively. RESULTS: A total of 179 patients with familial retinoblastoma were included. Paternal inheritance (PI) was identified in 109 (61%) patients and maternal inheritance (MI) in 70 patients (39%). A comparison (PI vs MI) revealed PI patients were older at presentation (57.2 vs 24.4 months [P = 0.002]) with no difference in patient sex (53% females vs 57% males [P = 0.606]) or number of family members affected (3.2 vs 3.0 family members [P = 0.255]). PI patients had more advanced classification according to the International Classification of Retinoblastoma (ICRB) (group E: 31% vs 8% [P = 0.012)] and greater largest tumor in basal diameter (9.0 vs 6.2 mm [P = 0.040]) and thickness (5.6 vs 4.0 mm [P = 0.038]); they were also less likely to be located in the macula (40% vs 60% [P = 0.004]). There was no difference in tumor laterality (69% vs 64% bilaterality [P = 0.530]). PI patients required enucleation more frequently (34% vs 14% [P = 0.007]). There was no difference in need for plaque radiotherapy (P = 0.86) or chemotherapy (P = 0.85). One PI patient developed metastatic retinoblastoma, and there were no retinoblastoma-related deaths. CONCLUSIONS: Patients with paternally inherited retinoblastoma presented at an older age, with larger, more peripheral tumors and more advanced ICRB group, and were more likely to require enucleation compared to those with maternally inherited retinoblastoma.

Improved Performance of ChatGPT-4 on the OKAP Examination: A Comparative Study with ChatGPT-3.5.

Introduction: This study aims to evaluate the performance of ChatGPT-4, an advanced artificial intelligence (AI) language model, on the Ophthalmology Knowledge Assessment Program (OKAP) examination compared to its predecessor, ChatGPT-3.5. Methods: Both models were tested on 180 OKAP practice questions covering various ophthalmology subject categories. Results: ChatGPT-4 significantly outperformed ChatGPT-3.5 (81% vs. 57%; p <0.001), indicating improvements in medical knowledge assessment. Discussion: The superior performance of ChatGPT-4 suggests potential applicability in ophthalmologic education and clinical decision support systems. Future research should focus on refining AI models, ensuring a balanced representation of fundamental and specialized knowledge, and determining the optimal method of integrating AI into medical education and practice.

The potential use of non-fungible tokens (NFTs) in healthcare and medical research.

Non-fungible tokens (NFTs) are cryptographic assets recorded on the blockchain that can certify authenticity and ownership, and they can be used to monetize health data, optimize the process of receiving a hematopoietic stem cell transplant, and improve the distribution of solid organs for transplantation. Blockchain technology, including NFTs, provides equitable access to wealth, increases transparency, eliminates personal or institutional biases of intermediaries, reduces inefficiencies, and ensures accountability. Blockchain architecture is ideal for ensuring security and privacy while granting individuals jurisdiction over their own information, making it a unique solution to the current limitations of existing health information systems. NFTs can be used to give patients the option to monetize their health data and provide valuable data to researchers. Wearable technology companies can also give their customers the option to monetize their data while providing data necessary to improve their products. Additionally, the process of receiving a hematopoietic stem cell transplant and the distribution of solid organs for transplantation could benefit from the integration of NFTs into the allocation process. However, there are limitations to the technology, including high energy consumption and the need for regulatory guidance. Further research is necessary to fully understand the potential of NFTs in healthcare and how it can be integrated with existing health information technology. Overall, NFTs have the potential to revolutionize the healthcare sector, providing benefits such as improved access to health information and increased efficiency in the distribution of organs for transplantation.

Retinal blood flow speed quantification at the capillary level using temporal autocorrelation fitting OCTA [Invited].

Optical coherence tomography angiography (OCTA) can visualize vasculature structures, but provides limited information about blood flow speed. Here, we present a second generation variable interscan time analysis (VISTA) OCTA, which evaluates a quantitative surrogate marker for blood flow speed in vasculature. At the capillary level, spatially compiled OCTA and a simple temporal autocorrelation model, ρ(τ) = exp(-ατ), were used to evaluate a temporal autocorrelation decay constant, α, as the blood flow speed marker. A 600 kHz A-scan rate swept-source OCT prototype instrument provides short interscan time OCTA and fine A-scan spacing acquisition, while maintaining multi mm2 field of views for human retinal imaging. We demonstrate the cardiac pulsatility and assess repeatability of α measured with VISTA. We show different α for different retinal capillary plexuses in healthy eyes and present representative VISTA OCTA in eyes with diabetic retinopathy.

The impact of commercial health datasets on medical research and health-care algorithms.

As the health-care industry emerges into a new era of digital health driven by cloud data storage, distributed computing, and machine learning, health-care data have become a premium commodity with value for private and public entities. Current frameworks of health data collection and distribution, whether from industry, academia, or government institutions, are imperfect and do not allow researchers to leverage the full potential of downstream analytical efforts. In this Health Policy paper, we review the current landscape of commercial health data vendors, with special emphasis on the sources of their data, challenges associated with data reproducibility and generalisability, and ethical considerations for data vending. We argue for sustainable approaches to curating open-source health data to enable global populations to be included in the biomedical research community. However, to fully implement these approaches, key stakeholders should come together to make health-care datasets increasingly accessible, inclusive, and representative, while balancing the privacy and rights of individuals whose data are being collected.