Search

Organoids, recognized as invaluable models in tumor and stem cell research, assume a pivotal role in the meticulous analysis of diverse datasets pertaining to their growth dynamics, drug screening processes and related phenomena. However, the manual scrutiny and conventional statistical methodologies employed in handling organoid data often grapple with challenges such as diminished precision and efficiency, heightened complexity, escalated human resource requirements, and a degree of subjectivity. Acknowledging the remarkable efficacy of artificial intelligence (AI) in the realms of biology and medicine, the incorporation of AI into organoid research stands poised to enhance the objectivity, precision and expediency of analyses. This integration empowers organoids to more effectively fulfill objectives such as disease modeling, drug screening and precision medicine. Notably, significant strides have been made in AI-driven analyses of organoid image data. The amalgamation of deep learning into image analysis facilitates a more meticulous delineation of the microstructural intricacies and nuanced changes within organoids, achieving a level of accuracy akin to that of experts. This not only elevates the precision of organoid morphology and growth recognition, but also contributes to substantial time and cost savings in research endeavors. Furthermore, the infusion of AI technology has yielded breakthroughs in the processing of organoid omics data, resulting in heightened efficiency in data processing and the identification of latent gene expression patterns. This furnishes novel tools for comprehending cellular development and unraveling the intricate mechanisms underlying various diseases. In addition to image data, AI techniques applied to diverse organoid datasets, encompassing electrical signals and spectra, have realized an unbiased classification of organoid types and states, embarking on a comprehensive journey towards characterizing organoids holistically. In the pivotal domain of drug screening for organoids, AI emerges as a stalwart companion, providing robust support for real-time process monitoring and result prediction. Leveraging high-content microscopy images and sophisticated deep learning models, researchers can dynamically monitor organoid responses to drugs, effecting non-invasive detection of drug impacts and amplifying the precision and efficiency of drug screening processes. Despite the significant strides made by AI in organoid research, challenges persist, encompassing hurdles in data acquisition, constraints in sample quality and quantity, and quandaries associated with model interpretability. Overcoming these challenges necessitates dedicated future research efforts aimed at enhancing data consistency, fortifying model interpretability, and exploring methodologies for the seamless fusion of multimodal data. Such endeavors are poised to usher in a more comprehensive and dependable application of AI in organoid research. In summation, the integration of AI technology introduces unparalleled opportunities to organoid research, resulting in noteworthy advancements. Nevertheless, interdisciplinary research and collaborative efforts remain imperative to navigate challenges and propel the more profound integration of AI into organoid research. The future holds promise for AI to assume an even more prominent role in advancing organoid research toward clinical translation and precision medicine.

Competing risks issues are common in clinical trials and epidemiological studies for patients in follow-up who may experience a variety of possible outcomes. Under such competing risks, two hazard-based statistical methods, cause-specific hazard (CSH) and subdistribution hazard (SDH), are frequently used to assess treatment effects among groups. However, the outcomes of the CSH-based and SDH-based methods have a close connection with the proportional hazards (CSH or SDH) assumption and may have an non-intuitive interpretation. Recently, restricted mean time lost (RMTL) has been used as an alternative summary measure for analysing competing risks, due to its clinical interpretability and robustness to the proportional hazards assumption. Considering the above approaches, we summarize the differences between hazard-based and RMTL-based methods from the aspects of practical interpretation, proportional hazards model assumption and the selection of restricted time points, and propose corresponding suggestions for the analysis of between-group differences under competing risks. Moreover, an R package 'cRMTL' and corresponding step-by-step guidance are available to help users for applying these approaches. [ABSTRACT FROM AUTHOR]

Background: In randomized controlled trials, multiple time‐to‐event endpoints are commonly used to determine treatment effects. However, choosing an appropriate method to address multiple endpoints, according to different purposes of clinical practice, is a challenge for researchers. Methods: We applied single endpoint, composite endpoint and win ratio analysis to chronic myeloid leukemia (CML) data to illustrate the distinctions with different multiple endpoints, including relapse, recovery and death after transplantation. Results: Regarding relapse and death, the hazard ratio in single endpoint analysis (HRs) were 1.281 (95% CI: 1.061−1.546) and hazard ratio in composite endpoint analysis (HRc) were 1.286 (95% CI: 1.112−1.486) and 1/WR (win ratio) was 1.292 (95% CI: 1.115−1.497) indicated a similar negative effect for non‐prophylaxis patients. However, when considering recovery and death, the corresponding HRs = 1.280 (95% CI: 1.056−1.552) may not be enough to describe the effect on death with nonproportional hazards (p < 0.05), and for the composite endpoint analysis, the HRc = 0.828 (95% CI: 0.740−0.926) cannot quantify and interpret the clinical effect on the composite endpoint with the combination of recovery and death, while the 1/WR = 1.351 (95% CI: 1.207−1.513) showed an unfavourable effect for non‐prophylaxis patients Conclusions: When dealing with multiple endpoints, single endpoints, researchers may choose single endpoints, composite endpoints and WR analysis due to different clinical applications and purposes. However, both single and composite endpoint analyses are hazard‐based measures, and thus, the proportional hazards assumption should be considered. Moreover, composite endpoint analysis should be applied for endpoints with similar clinical meanings but not opposing implications. Win ratio analysis can be considered for different clinical importance of multiple endpoints, but the meaning of 'winner' needs to be specified for desired or undesired endpoints. [ABSTRACT FROM AUTHOR]

In clinical and epidemiologic studies, hazard ratios are often applied to compare treatment effects between 2 groups for survival data. For competing-risks data, the corresponding quantities of interest are cause-specific hazard ratios and subdistribution hazard ratios. However, they both have some limitations related to model assumptions and clinical interpretation. Therefore, we recommend restricted mean time lost (RMTL) as an alternative measure that is easy to interpret in a competing-risks framework. Based on the difference in RMTL (RMTLd), we propose a new estimator, hypothetical test, and sample-size formula. Simulation results show that estimation of the RMTLd is accurate and that the RMTLd test has robust statistical performance (both type I error and statistical power). The results of 3 example analyses also verify the performance of the RMTLd test. From the perspectives of clinical interpretation, application conditions, and statistical performance, we recommend that the RMTLd be reported along with the hazard ratio in analyses of competing-risks data and that the RMTLd even be regarded as the primary outcome when the proportional hazards assumption fails. [ABSTRACT FROM AUTHOR]