We developed a deep learning radiomics nomogram (DLRN) using CT scans to improve clinical decision-making and risk stratification for early recurrence of hepatocellular carcinoma (HCC) after transplantation, which typically has a poor prognosis.

The Banff Liver Working Group recently published consensus recommendations for steatosis assessment in donor liver biopsy, but few studies reported their use and no automated deep-learning algorithms based on the proposed criteria have been developed so far. We evaluated Banff recommendations on a large monocentric series of donor liver needle biopsies by comparing pathologists' scores with those generated by convolutional neural networks (CNNs) we specifically developed for automated steatosis assessment.We retrospectively retrieved 292 allograft liver needle biopsies collected between January 2016 and January 2020 and performed steatosis assessment using a former intra-institution method (pre-Banff method) and the newly introduced Banff recommendations. Scores provided by pathologists and CNN models were then compared, and the degree of agreement was measured with the intraclass correlation coefficient (ICC).Regarding the pre-Banff method, poor agreement was observed between the pathologist and CNN models for small droplet macrovesicular steatosis (ICC: 0.38), large droplet macrovesicular steatosis (ICC: 0.08), and the final combined score (ICC: 0.16) evaluation, but none of these reached statistically significance. Interestingly, significantly improved agreement was observed using the Banff approach: ICC was 0.93 for the low-power score (p <0.001), 0.89 for the high-power score (p <0.001), and 0.93 for the final score (p <0.001). Comparing the pre-Banff method with the Banff approach on the same biopsy, pathologist and CNN model assessment showed a mean (±SD) percentage of discrepancy of 26.89 (±22.16) and 1.20 (±5.58), respectively.Our findings support the use of Banff recommendations in daily practice and highlight the need for a granular analysis of their effect on liver transplantation outcomes.We developed and validated the first automated deep-learning algorithms for standardized steatosis assessment based on the Banff Liver Working Group consensus recommendations. Our algorithm provides an unbiased automated evaluation of steatosis, which will lay the groundwork for granular analysis of steatosis's short- and long-term effects on organ viability, enabling the identification of clinically relevant steatosis cut-offs for donor organ acceptance. Implementing our algorithm in daily clinical practice will allow for a more efficient and safe allocation of donor organs, improving the post-transplant outcomes of patients.Copyright © 2023 The Author(s). Published by Elsevier B.V. All rights reserved.

Renal normothermic machine perfusion (NMP) is an organ preservation method based on the circulation of a warm (35-37 °C) perfusion solution through the renal vasculature to deliver oxygen and nutrients. However, its biological effects on marginal kidneys are unclear. We therefore used mass spectrometry to determine the proteomic profile of kidney tissue and urine from eight organs reconditioned for 120 min using a Kidney Assist device. Biopsies were taken during the pre-implantation histological evaluation (T-1), at the start of back table preparation (T0), and after 60 and 120 min of perfusion (T60, T120). Urine samples were collected at T0 (urine produced in the first 15 min after the beginning of normothermic reperfusion), T30, T60 and T120. Multiple algorithms, support vector machine learning and partial least squares discriminant analysis were used to select the most discriminative proteins during NMP. Statistical analysis revealed the upregulation of 169 proteins and the downregulation of 196 during NMP. Machine learning algorithms identified the top 50 most discriminative proteins, five of which were concomitantly upregulated (LXN, ETFB, NUDT3, CYCS and UQCRC1) and six downregulated (CFHR3, C1S, CFI, KNG1, SERPINC1 and F9) in the kidney and urine after NMP. Latexin (LXN), an endogenous carboxypeptidase inhibitor, resulted the most-upregulated protein at T120, and this result was confirmed by ELISA. In addition, functional analysis revealed that the most strongly upregulated proteins were involved in the oxidative phosphorylation system and ATP synthesis, whereas the downregulated proteins represented the complement system and coagulation cascade. Our proteomic analysis demonstrated that even brief periods of NMP induce remarkable metabolic and biochemical changes in marginal organs, which supports the use of this promising technique in the clinic.© 2023. The Author(s).

Medical digital twins, which are potentially vital for personalized medicine, have become a recent focus in medical research. Here we present an overview of the state of the art in medical digital twin development, especially in oncology and cardiology, where it is most advanced. We discuss major challenges, such as data integration and privacy, and provide an outlook on future advancements. Emphasizing the importance of this technology in healthcare, we highlight the potential for substantial improvements in patient-specific treatments and diagnostics.© 2024. Springer Nature America, Inc.

With biotechnological advancements, innovative omics technologies are constantly emerging that have enabled researchers to access multi-layer information from the genome, epigenome, transcriptome, proteome, metabolome, and more. A wealth of omics technologies, including bulk and single-cell omics approaches, have empowered to characterize different molecular layers at unprecedented scale and resolution, providing a holistic view of tumor behavior. Multi-omics analysis allows systematic interrogation of various molecular information at each biological layer while posing tricky challenges regarding how to extract valuable insights from the exponentially increasing amount of multi-omics data. Therefore, efficient algorithms are needed to reduce the dimensionality of the data while simultaneously dissecting the mysteries behind the complex biological processes of cancer. Artificial intelligence has demonstrated the ability to analyze complementary multi-modal data streams within the oncology realm. The coincident development of multi-omics technologies and artificial intelligence algorithms has fuelled the development of cancer precision medicine. Here, we present state-of-the-art omics technologies and outline a roadmap of multi-omics integration analysis using an artificial intelligence strategy. The advances made using artificial intelligence-based multi-omics approaches are described, especially concerning early cancer screening, diagnosis, response assessment, and prognosis prediction. Finally, we discuss the challenges faced in multi-omics analysis, along with tentative future trends in this field. With the increasing application of artificial intelligence in multi-omics analysis, we anticipate a shifting paradigm in precision medicine becoming driven by artificial intelligence-based multi-omics technologies.Copyright © 2023 Elsevier Ltd. All rights reserved.

Pancreatic surgery is considered one of the most challenging interventions by many surgeons, mainly due to retroperitoneal location and proximity to key and delicate vascular structures. These factors make pancreatic resection a demanding procedure, with successful rates far from optimal and frequent postoperative complications. Surgical planning is essential to improve patient outcomes, and in this regard, many technological advances made in the last few years have proven to be extremely useful in medical fields. This review aims to outline the potential and limitations of 3D digital and 3D printed models in pancreatic surgical planning, as well as the impact and challenges of novel technologies such as augmented/virtual reality systems or artificial intelligence to improve medical training and surgical outcomes.

The aim of this study was to evaluate a novel holographic craniofacial surgical planning application and its implementation throughout the planning and operative stages of facial transplantation by performing a critical analysis of comparative utility, cost, and limitations of MR and 3D printing.

Robotic-assisted kidney transplantation (RAKT) has shown solid results as a minimally invasive alternative to the standard open approach (open kidney transplantation [OKT]). However, RAKT is still limited in those cases where the recipient's iliac vessels present atherosclerotic plaques, frequently found in elder patients and in those subjected to long-term hemodialysis. Unlike OKT, where the surgeon can palpate the arterial plaques, in minimally invasive surgery the haptic feedback is missing, making the vascular clamping and arteriotomy unsafe.To employ three-dimensional (3D) imaging reconstruction using augmented reality (AR) to intraoperatively locate the plaques during the crucial steps of kidney transplantation.Our study was conducted according to the Idea, Development, Exploration, Assessment, and Long-term follow-up (IDEAL) model for surgical innovation. Three-dimensional virtual models were obtained from high-accuracy computed tomography scan imaging and superimposed on the vessels during RAKT using the Da Vinci console software.Three-dimensional AR-guided robotic-assisted kidney transplantation.The correspondence of virtual models with the real anatomy of patients was assessed comparing vessels' and plaques' measures.We tested the possibility of using the AR in the setting of vascular surgery by checking the correspondence of the virtual models to the real vessels. During the accuracy assessment, we investigated the anatomy of the iliac plaques and the capacity of the virtual models to correctly represent them. Finally, we tested the efficacy of the virtual model superimposition on the real vessels with plaques during RAKT in the recipients of living donor grafts. The main limitation consists in training needed to correctly superimpose virtual models on the real field.The employment of 3D AR allowed surgeons to overcome one of the main limitations of RAKT, setting the foundation to expand its indications to patients with advanced atheromatic vascular disease.The use of three-dimensional augmented reality guidance during kidney transplantation (KT) has the potential to "navigate" the surgeon during KT, allowing a safer procedure in patients with atheromatic vascular disease.Copyright © 2022 European Association of Urology. Published by Elsevier B.V. All rights reserved.

Artificial intelligence (AI) is rapidly emerging in healthcare, yet applications in surgery remain relatively nascent. Here we review the integration of AI in the field of surgery, centering our discussion on multifaceted improvements in surgical care in the preoperative, intraoperative and postoperative space. The emergence of foundation model architectures, wearable technologies and improving surgical data infrastructures is enabling rapid advances in AI interventions and utility. We discuss how maturing AI methods hold the potential to improve patient outcomes, facilitate surgical education and optimize surgical care. We review the current applications of deep learning approaches and outline a vision for future advances through multimodal foundation models.© 2024. Springer Nature America, Inc.

Robot assisted surgery (RAS) provides medical practitioners with valuable tools, decreasing strain during surgery and leading to better patient outcomes. While the loss of haptic sensation is a commonly cited disadvantage of RAS, new systems aim to address this problem by providing artificial haptic feedback. N = 56 papers that compared robotic surgery systems with and without haptic feedback were analyzed to quantify the performance benefits of restoring the haptic modality. Additionally, this study identifies factors moderating the effect of restoring haptic sensation. Overall results showed haptic feedback was effective in reducing average forces (Hedges’ g = 0.83) and peak forces (Hedges’ g = 0.69) applied during surgery, as well as reducing the completion time (Hedges’ g = 0.83). Haptic feedback has also been found to lead to higher accuracy (Hedges’ g = 1.50) and success rates (Hedges’ g = 0.80) during surgical tasks. Effect sizes on several measures varied between tasks, the type of provided feedback, and the subjects’ levels of surgical expertise, with higher levels of expertise generally associated with smaller effect sizes. No significant differences were found between virtual fixtures and rendering contact forces. Implications for future research are discussed.

Donor-derived cell-free DNA (dd-cfDNA) is an emerging non-invasive biomarker that has the potential to detect allograft injury. The capacity of donor-derived cell-free DNA to detect kidney allograft rejection and its added clinical value beyond standard of care patient monitoring is unclear. We enrolled 2,882 kidney allograft recipients from 14 transplantation centers in Europe and the US in an observational population-based study. The primary analysis included 1,134 patients. Donor-derived cell-free DNA levels strongly correlated with allograft rejection including antibody-mediated rejection (p < 0.0001), T-Cell mediated rejection (p < 0.0001) and mixed rejection (p < 0.0001). In multivariable analysis, circulating dd-cfDNA was significantly associated with allograft rejection (OR: 2.275; CI:1.902-2.739; p < 0.0001) independently of standard of care patient monitoring parameters. The inclusion of dd-cfDNA to a standard of care prediction model showed improved discrimination (0.777 [95% CI: 0.741-0.811] to 0.821 [95% CI: 0.784-0.852]; p = 0.0011) and calibration. These results were confirmed in the external validation cohorts (n = 1,748) including a cohort of African American patients (n = 439). Finally, dd-cfDNA showed high predictive value to detect subclinical rejection in stable patients. Our study provides insights on the potential value of assessing dd-cfDNA, in addition to standard of care monitoring, to improve the detection of allograft rejection. ClinicalTrials.gov registration: NCT05995379.© 2024. The Author(s), under exclusive licence to Springer Nature America, Inc.

Liquid biopsies (LBs), particularly using circulating tumor DNA (ctDNA), are expected to revolutionize precision oncology and blood-based cancer screening. Recent technological improvements, in combination with the ever-growing understanding of cell-free DNA (cfDNA) biology, are enabling the detection of tumor-specific changes with extremely high resolution and new analysis concepts beyond genetic alterations, including methylomics, fragmentomics, and nucleosomics. The interrogation of a large number of markers and the high complexity of data render traditional correlation methods insufficient. In this regard, machine learning (ML) algorithms are increasingly being used to decipher disease- and tissue-specific signals from cfDNA. Here, we review recent insights into biological ctDNA features and how these are incorporated into sophisticated ML applications.Copyright © 2023 The Authors. Published by Elsevier Ltd.. All rights reserved.

Drug development is a complex and time-consuming endeavor that traditionally relies on the experience of drug developers and trial-and-error experimentation. The advent of artificial intelligence (AI) technologies, particularly emerging large language models and generative AI, is poised to redefine this paradigm. The integration of AI-driven methodologies into the drug development pipeline has already heralded subtle yet meaningful enhancements in both the efficiency and effectiveness of this process. Here we present an overview of recent advancements in AI applications across the entire drug development workflow, encompassing the identification of disease targets, drug discovery, preclinical and clinical studies, and post-market surveillance. Lastly, we critically examine the prevailing challenges to highlight promising future research directions in AI-augmented drug development.© 2025. Springer Nature America, Inc.

Artificial Intelligence (AI) and machine learning (ML) are increasingly being applied across the transplantation care pathway, supporting tasks such as donor–recipient matching, immunological risk stratification, early detection of graft dysfunction, and optimisation of immunosuppressive therapy. This review provides a structured synthesis of current AI applications in transplantation, with a focus on underrepresented areas including real-time graft viability assessment, adaptive immunosuppression, and cross-organ immune modelling. The review also examines the translational infrastructure needed for clinical implementation, such as federated learning, explainable AI (XAI), and data governance. Evidence suggests that AI-based models can improve predictive accuracy and clinical decision support when compared to conventional approaches. However, limitations related to data quality, algorithmic bias, model transparency, and integration into clinical workflows remain. Addressing these challenges through rigorous validation, ethical oversight, and interdisciplinary collaboration will be necessary to support the safe and effective use of AI in transplant medicine.