Infections after surgery remain a leading cause of morbidity and mortality, yet reliable risk stratification at the end of surgery is limited. Intraoperative vital signs are continuously recorded in modern operating rooms but remain an underexploited source of real-time prognostic information. We developed and validated a machine-learning model integrating intraoperative vital-sign dynamics to predict postoperative infections immediately at the end of surgery. We extracted arterial blood pressure, heart rate, oxygen saturation, temperature, and end-tidal CO2 time-series from a clinical data warehouse, transforming these signals into interpretable summary, trend, and distributional descriptors. Using routine data from 10,719 surgical procedures, models incorporating interpretable intraoperative time-series features achieved an AUROC of 0.88 (95% CI, 0.85–0.91) for infection prediction at the end of surgery, significantly outperforming models based on preoperative variables alone. Model predictions were calibrated across major procedure clusters and interpretable through SHAP-based feature attribution. Our results demonstrate that intraoperative time-series data encode signatures of cumulative surgical and physiological stress, revealing early and clinically actionable signals of postoperative infection risk and enable an explainable machine-learning framework for perioperative monitoring systems.
@article{blatter_end-of-surgery_2026,title={End-of-surgery prediction of postoperative infectious complications from intraoperative vital-sign dynamics},issn={2398-6352},url={https://www.nature.com/articles/s41746-026-02707-1},doi={10.1038/s41746-026-02707-1},language={en},journal={npj Digital Medicine},author={Blatter, Tobias U. and Wintsch, Yves and Triep, Karen and Endrich, Olga and Guillen-Ramirez, Hugo and Beldi, Guido},month=may,year={2026},}
LLM-augmented semantic embeddings enable Cross-Lingual mapping of medical procedure terms
Hugo Guillen-Ramirez, Karen Triep, Christophe Gaudet-Blavignac, and
3 more authors
Cross-lingual information retrieval limits global exchange of data because of the high diversity in the methods to classify, document and encode medical procedures. Traditional keyword-based or single-language systems are not able to align data from surgical and interventional procedures, especially from non-English healthcare systems. This study aims to develop a pipeline for cross-lingual retrieval and integration of medical procedures data. MAP-CARE is a novel framework that leverages Large Language Models (LLMs) for translating and transforming medical procedures into a unified multilingual embedding space. Semantic embeddings are used to enhance retrieval accuracy and interoperability across languages and healthcare systems. MAP-CARE demonstrated high accuracy in the translation and mapping of clinical terms. Its cross-language translation performance proved robust, achieving up to Acc@5 = 0.90 in translating procedure classification codes across English, German, French, and Italian. The cross-classification mapping workflow also showed high accuracy in aligning two different national procedure classifications, with exact and near matches exceeding 53.8% at the most granular level. MAP-CARE offers a flexible, scalable, and robust solution for the multilingual and cross-system integration of medical procedural data. Its innovative use of large language models combined with semantic embeddings sets a new standard for the accessibility and utility of multilingual medical information. The framework is designed for easy extension from a terminology file in CSV format and is publicly available.
@article{guillen-ramirez_llm-augmented_2026,title={{LLM}-augmented semantic embeddings enable {Cross}-{Lingual} mapping of medical procedure terms},issn={2045-2322},url={https://www.nature.com/articles/s41598-025-34778-7},doi={10.1038/s41598-025-34778-7},language={en},urldate={2026-01-22},journal={Scientific Reports},author={Guillen-Ramirez, Hugo and Triep, Karen and Gaudet-Blavignac, Christophe and Phull, Baljit and Beldi, Guido and Endrich, Olga},month=jan,year={2026},}
2025
Prediction of postoperative infections by strategic data imputation and explainable machine learning
Hugo Guillen-Ramirez, Daniel Sanchez-Taltavull, Stéphanie Perrodin, and
5 more authors
Journal of the American Medical Informatics Association, Nov 2025
Objectives: Infections following healthcare-associated interventions drive patient morbidity and mortality, making early detection essential. Traditional predictive models utilize preoperative surgical characteristics. This study evaluated whether integrating postoperative laboratory values and their kinetics could improve outcome prediction. Materials and Methods: 91,794 surgical cases were extracted from electronic health records (EHR) and analyzed to predict bacterial infection as the endpoint. The endpoint was documented in the EHR as ICD-10 by a professional coding team. Variables were grouped as preoperative, intraoperative, or postoperative. Strategic imputation was used for postoperative missing laboratory values. Procedure-agnostic prediction models were built incorporating both static and kinetic properties of laboratory values. Results: The integration of kinetics of laboratory values into a machine learning predictor achieved a recall, precision and ROC AUC at postoperative day 2 of 0.71, 0.69, and 0.83, respectively. Moreover, infection detection outperformed clinician-based decision-making, as reflected by the postoperative timing of antibiotic administration. The analysis identified previously unknown, informative combinations of routine markers from hepatic, renal, and bone marrow functions that predict outcome. Discussion: Dynamic modelling of postoperative laboratory values enhanced the timeliness and accuracy of infection detection compared with static or preoperative-only models. The integration of explainable machine learning supports clinical interpretation and highlights the contribution of multiple organ systems to postoperative infection risk. Conclusion: A surgery-independent workflow integrating time-series values from laboratory parameters to enhance baseline predictors of infection. This interpretable approach is generalizable across procedures and has the potential to optimize patient outcomes and resource use in surgical care.
@article{guillen-ramirez_prediction_2025,title={Prediction of postoperative infections by strategic data imputation and explainable machine learning},volume={32},copyright={https://creativecommons.org/licenses/by/4.0/},issn={1067-5027, 1527-974X},url={https://academic.oup.com/jamia/article/32/11/1706/8244921},doi={10.1093/jamia/ocaf145},language={en},number={11},urldate={2026-01-22},journal={Journal of the American Medical Informatics Association},author={Guillen-Ramirez, Hugo and Sanchez-Taltavull, Daniel and Perrodin, Stéphanie and Peisl, Sarah and Triep, Karen and Gaudet-Blavignac, Christophe and Endrich, Olga and Beldi, Guido},month=nov,year={2025},pages={1706--1717},}
Preoperative Enterosignatures Predict Surgical Site Infections After Abdominal Surgery
Simone N Zwicky, Daniel Spari, Daniel Rodjakovic, and
3 more authors
@article{zwicky_preoperative_2025,title={Preoperative {Enterosignatures} {Predict} {Surgical} {Site} {Infections} {After} {Abdominal} {Surgery}},volume={12},copyright={https://creativecommons.org/licenses/by/4.0/},issn={2328-8957},url={https://academic.oup.com/ofid/article/doi/10.1093/ofid/ofaf549/8246400},doi={10.1093/ofid/ofaf549},language={en},number={9},urldate={2026-01-22},journal={Open Forum Infectious Diseases},author={Zwicky, Simone N and Spari, Daniel and Rodjakovic, Daniel and Guillen-Ramirez, Hugo and Yilmaz, Bahtiyar and Beldi, Guido},month=aug,year={2025},pages={ofaf549},}
2024
Influence of patient characteristics on microbial composition in surgical-site infections: insights from national surveillance study
Sarah Peisl, Hugo Guillen-Ramirez, Daniel Sánchez-Taltavull, and
3 more authors
@article{peisl_influence_2024,title={Influence of patient characteristics on microbial composition in surgical-site infections: insights from national surveillance study},volume={111},issn={1365-2168},shorttitle={Influence of patient characteristics on microbial composition in surgical-site infections},doi={10.1093/bjs/znae138},language={eng},number={6},journal={The British Journal of Surgery},author={Peisl, Sarah and Guillen-Ramirez, Hugo and Sánchez-Taltavull, Daniel and Widmer, Andreas and Sommerstein, Rami and Beldi, Guido},month=jun,year={2024},pmid={38926136},keywords={Adult, Aged, Female, Humans, Male, Middle Aged, Risk Factors, Antibiotic Prophylaxis, Surgical Wound Infection, Retrospective Studies, Age Factors, Body Mass Index, Operative Time, Switzerland},pages={znae138},}
Functional identification of \textitcis -regulatory long noncoding RNAs at controlled false discovery rates
Bhavya Dhaka, Marc Zimmerli, Daniel Hanhart, and
12 more authors
@article{dhaka_functional_2024,title={Functional identification of \textit{cis} -regulatory long noncoding {RNAs} at controlled false discovery rates},volume={52},issn={0305-1048, 1362-4962},url={https://academic.oup.com/nar/article/52/6/2821/7606963},doi={10.1093/nar/gkae075},language={en},number={6},urldate={2026-01-22},journal={Nucleic Acids Research},author={Dhaka, Bhavya and Zimmerli, Marc and Hanhart, Daniel and Moser, Mario B and Guillen-Ramirez, Hugo and Mishra, Sanat and Esposito, Roberta and Polidori, Taisia and Widmer, Maro and García-Pérez, Raquel and Julio, Marianna Kruithof-de and Pervouchine, Dmitri and Melé, Marta and Chouvardas, Panagiotis and Johnson, Rory},month=apr,year={2024},pages={2821--2835},}
Noise in the operating room coincides with surgical difficulty
Sarah Peisl, Daniel Sánchez-Taltavull, Hugo Guillen-Ramirez, and
10 more authors
Noise in the operating room has been shown to distract the surgical team and to be associated with postoperative complications. It is, however, unclear whether complications after noisy operations are the result of objective or subjective surgical difficulty or the consequence of distraction of the operating room team by noise. Noise level measurements were prospectively performed during operations in four Swiss hospitals. Objective difficulty for each operation was calculated based on surgical magnitude as suggested by the Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity (POSSUM), duration of operation and surgical approach. Subjective difficulty and distraction were evaluated by a questionnaire filled out by the operating room team members. Complications were assessed 30 days after surgery. Multivariable regression analyses revealed that subjective difficulty as reported by all members of the surgical team, but not distraction, was highly associated with noise and complications. Only objective surgical difficulty independently predicted noise and postoperative complications. Noise in the operating room is a surrogate of surgical difficulty and thereby predicts postoperative complications.
@article{peisl_noise_2024,title={Noise in the operating room coincides with surgical difficulty},volume={8},issn={2474-9842},url={https://academic.oup.com/bjsopen/article/8/5/zrae098/7823788},doi={10.1093/bjsopen/zrae098},language={en},number={5},journal={BJS Open},author={Peisl, Sarah and Sánchez-Taltavull, Daniel and Guillen-Ramirez, Hugo and Tschan, Franziska and Semmer, Norbert K. and Hübner, Martin and Demartines, Nicolas and Wrann, Simon G. and Gutknecht, Stefan and Weber, Markus and Candinas, Daniel and Beldi, Guido and Keller, Sandra},month=oct,year={2024},pages={zrae098},}
2023
Tumour mutations in long noncoding RNAs enhance cell fitness
Roberta Esposito, Andrés Lanzós, Tina Uroda, and
30 more authors
@article{esposito_tumour_2023,title={Tumour mutations in long noncoding {RNAs} enhance cell fitness},volume={14},issn={2041-1723},url={https://www.nature.com/articles/s41467-023-39160-7},doi={10.1038/s41467-023-39160-7},language={en},number={1},urldate={2026-01-22},journal={Nature Communications},author={Esposito, Roberta and Lanzós, Andrés and Uroda, Tina and Ramnarayanan, Sunandini and Büchi, Isabel and Polidori, Taisia and Guillen-Ramirez, Hugo and Mihaljevic, Ante and Merlin, Bernard Mefi and Mela, Lia and Zoni, Eugenio and Hovhannisyan, Lusine and McCluggage, Finn and Medo, Matúš and Basile, Giulia and Meise, Dominik F. and Zwyssig, Sandra and Wenger, Corina and Schwarz, Kyriakos and Vancura, Adrienne and Bosch-Guiteras, Núria and Andrades, Álvaro and Tham, Ai Ming and Roemmele, Michaela and Medina, Pedro P. and Ochsenbein, Adrian F. and Riether, Carsten and Kruithof-de Julio, Marianna and Zimmer, Yitzhak and Medová, Michaela and Stroka, Deborah and Fox, Archa and Johnson, Rory},month=jun,year={2023},pages={3342},}
2022
Multi-hallmark long noncoding RNA maps reveal non-small cell lung cancer vulnerabilities
Roberta Esposito, Taisia Polidori, Dominik F. Meise, and
26 more authors
@article{esposito_multi-hallmark_2022,title={Multi-hallmark long noncoding {RNA} maps reveal non-small cell lung cancer vulnerabilities},volume={2},issn={2666979X},url={https://linkinghub.elsevier.com/retrieve/pii/S2666979X22001136},doi={10.1016/j.xgen.2022.100171},language={en},number={9},urldate={2023-05-06},journal={Cell Genomics},author={Esposito, Roberta and Polidori, Taisia and Meise, Dominik F. and Pulido-Quetglas, Carlos and Chouvardas, Panagiotis and Forster, Stefan and Schaerer, Paulina and Kobel, Andrea and Schlatter, Juliette and Kerkhof, Erik and Roemmele, Michaela and Rice, Emily S. and Zhu, Lina and Lanzós, Andrés and Guillen-Ramirez, Hugo A. and Basile, Giulia and Carrozzo, Irene and Vancura, Adrienne and Ullrich, Sebastian and Andrades, Alvaro and Harvey, Dylan and Medina, Pedro P. and Ma, Patrick C. and Haefliger, Simon and Wang, Xin and Martinez, Ivan and Ochsenbein, Adrian F. and Riether, Carsten and Johnson, Rory},month=sep,year={2022},pages={100171},}
2021
Enhancing CRISPR deletion via pharmacological delay of DNA-PKcs
Núria Bosch-Guiteras, Tina Uroda, Hugo A. Guillen-Ramirez, and
7 more authors
@article{bosch-guiteras_enhancing_2021,title={Enhancing {CRISPR} deletion via pharmacological delay of {DNA}-{PKcs}},volume={31},issn={1088-9051, 1549-5469},url={http://genome.cshlp.org/lookup/doi/10.1101/gr.265736.120},doi={10.1101/gr.265736.120},language={en},number={3},urldate={2023-05-06},journal={Genome Research},author={Bosch-Guiteras, Núria and Uroda, Tina and Guillen-Ramirez, Hugo A. and Riedo, Rahel and Gazdhar, Amiq and Esposito, Roberta and Pulido-Quetglas, Carlos and Zimmer, Yitzhak and Medová, Michaela and Johnson, Rory},month=mar,year={2021},pages={461--471},}
2020
Automatic construction of molecular similarity networks for visual graph mining in chemical space of bioactive peptides: an unsupervised learning approach
Longendri Aguilera-Mendoza, Yovani Marrero-Ponce, César R. García-Jacas, and
4 more authors
@article{aguilera-mendoza_automatic_2020,title={Automatic construction of molecular similarity networks for visual graph mining in chemical space of bioactive peptides: an unsupervised learning approach},volume={10},issn={2045-2322},shorttitle={Automatic construction of molecular similarity networks for visual graph mining in chemical space of bioactive peptides},url={https://www.nature.com/articles/s41598-020-75029-1},doi={10.1038/s41598-020-75029-1},language={en},number={1},urldate={2023-05-06},journal={Scientific Reports},author={Aguilera-Mendoza, Longendri and Marrero-Ponce, Yovani and García-Jacas, César R. and Chavez, Edgar and Beltran, Jesus A. and Guillen-Ramirez, Hugo A. and Brizuela, Carlos A.},month=oct,year={2020},pages={18074},}
2019
Graph-based data integration from bioactive peptide databases of pharmaceutical interest: toward an organized collection enabling visual network analysis
Longendri Aguilera-Mendoza, Yovani Marrero-Ponce, Jesus A Beltran, and
3 more authors
@article{aguilera-mendoza_graph-based_2019,title={Graph-based data integration from bioactive peptide databases of pharmaceutical interest: toward an organized collection enabling visual network analysis},volume={35},issn={1367-4803, 1367-4811},shorttitle={Graph-based data integration from bioactive peptide databases of pharmaceutical interest},url={https://academic.oup.com/bioinformatics/article/35/22/4739/5474901},doi={10.1093/bioinformatics/btz260},language={en},number={22},urldate={2023-05-06},journal={Bioinformatics},author={Aguilera-Mendoza, Longendri and Marrero-Ponce, Yovani and Beltran, Jesus A and Tellez Ibarra, Roberto and Guillen-Ramirez, Hugo A and Brizuela, Carlos A},editor={Wren, Jonathan},month=nov,year={2019},pages={4739--4747},}
Particle size distribution from extinction and absorption data of metallic nanoparticles
J. Gabriela Calvillo-Vázquez, Hugo A. Guillén-Ramírez, Melissa DiazDuarte-Rodríguez, and
2 more authors
@article{calvillo-vazquez_particle_2019,title={Particle size distribution from extinction and absorption data of metallic nanoparticles},volume={58},issn={1559-128X, 2155-3165},doi={10.1364/AO.58.009955},language={en},number={36},urldate={2023-05-06},journal={Applied Optics},author={Calvillo-Vázquez, J. Gabriela and Guillén-Ramírez, Hugo A. and DiazDuarte-Rodríguez, Melissa and Licea-Claverie, Angel and Méndez, Eugenio R.},month=dec,year={2019},pages={9955},}
2018
Classification of riboswitch sequences using k-mer frequencies
Hugo A. Guillén-Ramírez, and Israel M. Martínez-Pérez
Riboswitches are non-coding RNAs that regulate gene expression by altering the structural conformation of mRNA transcripts. Their regulation mechanism might be exploited for interesting biomedical applications such as drug targets and biosensors. A major challenge consists in accurately identifying metabolite-binding RNA switches which are structurally complex and diverse. In this regard, we investigated the classification of 16 riboswitch families using supervised learning algorithms trained solely with sequence-based features. We generated a reduced feature set and proposed a visual representation to explore its components. We induced Support Vector Machine, Random Forest, Naive Bayes, J48, and HyperPipes classifiers with our proposed feature set and tested their performance over independent data. Our best multi-class classifier achieved F-measure values of 0.996 and 0.966 in the training and test phases, respectively, outperforming those of a previous approach. When compared against BLAST, our best classifiers yielded competitive results. This work shows that the classifiers trained with our sequence-based feature set accurately discriminate riboswitches.
@article{guillen-ramirez_classification_2018,title={Classification of riboswitch sequences using k-mer frequencies},volume={174},issn={03032647},url={https://linkinghub.elsevier.com/retrieve/pii/S0303264718302077},doi={10.1016/j.biosystems.2018.09.001},language={en},urldate={2023-05-06},journal={Biosystems},author={Guillén-Ramírez, Hugo A. and Martínez-Pérez, Israel M.},month=dec,year={2018},pages={63--76},}
A benchmark of heart sound classification systems based on sparse decompositions
Roilhi Frajo Ibarra-Hernández, Nancy Bertin, Miguel Angel Alonso-Arévalo, and
1 more author
In 14th International Symposium on Medical Information Processing and Analysis, Dec 2018
@inproceedings{ibarra-hernandez_benchmark_2018,address={Mazatlán, Mexico},title={A benchmark of heart sound classification systems based on sparse decompositions},isbn={978-1-5106-2605-8 978-1-5106-2606-5},url={https://www.spiedigitallibrary.org/conference-proceedings-of-spie/10975/2506758/A-benchmark-of-heart-sound-classification-systems-based-on-sparse/10.1117/12.2506758.full},doi={10.1117/12.2506758},urldate={2023-05-06},booktitle={14th {International} {Symposium} on {Medical} {Information} {Processing} and {Analysis}},publisher={SPIE},author={Ibarra-Hernández, Roilhi Frajo and Bertin, Nancy and Alonso-Arévalo, Miguel Angel and Guillén-Ramírez, Hugo Armando},editor={Romero, Eduardo and Lepore, Natasha and Brieva, Jorge},month=dec,year={2018},pages={14},}
2017
Accurate classification of immunomodulatory RNA sequences
Hugo A. Guillen-Ramirez, Jose Colbes, Carlos A. Brizuela, and
1 more author
In 2017 International Joint Conference on Neural Networks (IJCNN), May 2017
@inproceedings{guillen-ramirez_accurate_2017,address={Anchorage, AK, USA},title={Accurate classification of immunomodulatory {RNA} sequences},isbn={978-1-5090-6182-2},url={http://ieeexplore.ieee.org/document/7965860/},doi={10.1109/IJCNN.2017.7965860},urldate={2023-05-06},booktitle={2017 {International} {Joint} {Conference} on {Neural} {Networks} ({IJCNN})},publisher={IEEE},author={Guillen-Ramirez, Hugo A. and Colbes, Jose and Brizuela, Carlos A. and Martinez-Perez, Israel M.},month=may,year={2017},pages={236--241},}
2013
Improving an evolutionary multi-objective algorithm for the biclustering of gene expression data
Carlos A. Brizuela, Jorge E. Luna-Taylor, Israel Martinez-Perez, and
3 more authors
In 2013 IEEE Congress on Evolutionary Computation, Jun 2013
@inproceedings{brizuela_improving_2013,address={Cancun, Mexico},title={Improving an evolutionary multi-objective algorithm for the biclustering of gene expression data},isbn={978-1-4799-0454-9 978-1-4799-0453-2 978-1-4799-0451-8 978-1-4799-0452-5},url={http://ieeexplore.ieee.org/document/6557574/},doi={10.1109/CEC.2013.6557574},urldate={2023-05-06},booktitle={2013 {IEEE} {Congress} on {Evolutionary} {Computation}},publisher={IEEE},author={Brizuela, Carlos A. and Luna-Taylor, Jorge E. and Martinez-Perez, Israel and Guillen, Hugo A. and Rodriguez, David O. and Beltran-Verdugo, Armando},month=jun,year={2013},pages={221--228},}