Title correction sharpens the record for a kidney transplant prediction study
PLOS ONE issued a correction to a 2026 article on weight gain after kidney transplantation, fixing the title and citation for a paper centered on the feasibility of a machine learning model in this setting. The revised title now reads: “Feasibility of a machine learning model for predicting weight gain after kidney transplantation.” The correction also amends the citation details tied to the original publication, preserving the bibliographic record for future indexing, retrieval, and citation tracking.
A correction of this kind may look editorial on the surface, but in transplant medicine precision is not cosmetic. Weight trajectories after transplantation influence metabolic risk, cardiovascular burden, and long-term graft care, so a study aimed at prediction needs to be labeled with exacting clarity. When the title mistakenly repeats or distorts key wording, readers can misinterpret the scope of the work before they ever reach the methods section.
Why accurate naming matters in transplant prediction research
Kidney transplantation changes the clinical calculus for many patients. Immunosuppression, recovery from uremia, improved appetite, and changes in activity level can all alter body weight after surgery. Some of those changes are expected; others may become clinically meaningful if weight gain compounds hypertension, dyslipidemia, insulin resistance, or reduced mobility. A model designed to anticipate which recipients are most likely to gain weight can help clinicians think earlier about nutrition, counseling, and follow-up intensity.
Machine learning is particularly attractive in this space because transplant recipients often generate dense longitudinal data. Laboratory values, medication exposures, comorbidities, and early postoperative trends can interact in ways that are difficult to capture with a single linear rule. In principle, a model can integrate those features and estimate risk in a way that supports individualized surveillance. The correction does not provide the model’s variables, performance, or validation strategy, and those details should not be inferred. What it does establish is the study’s intended focus: feasibility, not certainty.
Feasibility studies occupy an important place in clinical data science. They ask whether a predictive approach can be built, whether the available data are usable, and whether the target outcome is defined closely enough to support future development. In a transplant cohort, that step matters because weight gain is not just a number on a scale; it may reflect a mixture of steroid exposure, dietary shifts, fluid status, and lifestyle recovery. A model that cannot distinguish those contributors risks producing predictions that are technically elegant but clinically thin.
A small bibliographic correction with practical downstream effects
The corrected citation now lists the article as: de Freitas JdMC, Ballester PL, Pedrollo EF, Corrêa C, Menezes RCR, Leitão CB, et al. (2026) Feasibility of a machine learning model for predicting weight gain after kidney transplantation. PLoS One 21(10): e0360501. The correction itself was published in November 2026. Those details matter because citation databases, systematic reviewers, and transplant researchers rely on exact bibliographic metadata to link corrections with the underlying paper.
Mislabeling a title can create practical friction. Search engines may misfile the study, citation managers may duplicate it incorrectly, and readers searching for artificial intelligence work in transplantation may miss it entirely if the subject line is malformed. In a fast-moving area such as medical data science, even a small title error can distort discoverability. The correction restores alignment between the study title, the topic it addresses, and the formal record that journals use for indexing.
A corrected title also protects interpretation at the bedside. Clinicians reading only headlines or abstracts may assume a paper addresses treatment effects when it actually explores model feasibility. In transplantation, where postoperative weight gain has both metabolic and behavioral dimensions, that distinction shapes expectations. A feasibility analysis can justify the next step; it does not replace clinical judgment, nutritional monitoring, or prospective validation in a live population.
Prediction tools in transplantation still need clinical grounding
The broader promise of machine learning in kidney transplantation is straightforward: earlier risk recognition could allow more tailored intervention. If a recipient appears likely to gain excessive weight, clinicians may intensify nutrition counseling, medication review, and long-term metabolic monitoring. Such tools may also help researchers identify patterns that deserve prospective testing. Yet model development remains vulnerable to overfitting, dataset bias, and inconsistent outcome definitions, especially when the target is influenced by treatment effects and behavior at the same time.
Those limitations are exactly why a correction like this deserves attention. Clear labeling helps preserve the chain from data to clinical use. A model aimed at predicting weight gain after kidney transplantation belongs in a carefully defined evidence pathway: feasibility, validation, external testing, and then, only if the evidence holds, possible integration into routine transplant follow-up.
For now, the record is cleaner. The article’s corrected title states its purpose plainly, and the amended citation ensures that future readers searching the transplant literature will arrive at the right paper, not a mislabeled version of it. The remaining scientific questions are not about the correction itself, but about whether a machine learning approach can reliably predict post-transplant weight gain across different populations and care settings.