LHLukas Hilgendorf

Lukas Hilgendorf

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Lukas Hilgendorf has a total of 45 co-authors. Here are the top co-authors by number of shared publications:

  • Aidin Rawshani

    Institute of Medicine, Department of Molecular and Clinical Medicine, University of Gothenburg, Sahlgrenska Academy , Vita stråket 15, Sahlgrenska sjukhuset, Gothenburg 41345, Sweden

    3 shared publications
  • Deepak L. Bhatt

    Mount Sinai Fuster Heart Hospital, Icahn School of Medicine at Mount Sinai, New York, NY (D.L.B.).

    3 shared publications
  • Erik Andersson

    Department of Molecular and Clinical Medicine, Institute of Medicine, University of Gothenburg , Bruna stråket 16, 413 45 Göteborg, Gothenburg

    3 shared publications
  • Kristofer Skoglund

    Department of Cardiology, Sahlgrenska University Hospital , 41345 Gothenburg

    3 shared publications
  • Araz Rawshani

    Institute of Medicine, Department of Molecular and Clinical Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; Wallenberg Centre for Molecular and Translational Research (WCMTM), Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; Department of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden

    2 shared publications
  • Araz Rawshani

    Department of Molecular and Clinical Medicine, Institute of Medicine, University of Gothenburg, Gothenburg, Sweden

    2 shared publications

Latest Publications

  • Open Access

    Multi-modal deep-learning troponin prediction from electrocardiograms and demographic data

    by Lukas Hilgendorf, Pétur Pétursson, Erik Andersson, Aidin Rawshani, Deepak L Bhatt, Vibha Gupta, Peter Lundgren, Kristofer Skoglund, Truls Råmunddal, Araz Rawshani

    Aims: Electrocardiograms (ECGs) and troponin (Tn) testing are essential tools for the diagnosis and management of cardiac conditions. Prompt diagnosis using these tools can significantly improve patient outcomes. Methods and results: The objective of this study was to design and create a deep-learning model capable of predicting high-sensitivity troponin (hs-Tn) elevation in patients undergoing chest-pain triage. We developed a novel, multi-modal, externally validated deep-learning model that incorporates ECG data, age, and sex to predict high-sensitivity troponin-T elevation. The dataset used for this study was multi-centre and externally validated, drawing from data collected in two emergency rooms. The study population included all patients presenting to the ER with either chest pain or dyspnoea during the study period, where an ECG was recorded and a Tn test was performed. The model was trained on a dataset comprising 35 821 ECGs, with a positive fraction of 35.7%. It achieved an internal area under the receiver operating characteristic (AUROC) of 0.8958 ± 0.0040 (95% CI) and an AUROC of 0.8765 ± 0.0110 in external validation. The model's Score-CAM saliency maps demonstrated high activation from the ST-segment, indicating that the model draws information from relevant ECG segments. Conclusion: This study presents new opportunities for enhancing triage processes, enabling more rapid and accurate alerts to physicians regarding acute myocardial infarctions. The primary benefit of predicting Tn elevation lies in the objectivity of the label compared with compounded clinical outcomes and diagnoses.

  • Open Access

    Multimodal deep learning for acute myocardial infarction detection from 12-lead electrocardiogram: a multi-centre study with cross-hospital validation

    by Vibha Gupta, Lukas Hilgendorf, Erik Andersson, Antros Louca, Arman Shahmari, Alfred Hjalmarsson, Rajkumar Saini, Carlo Pirazzi, Monér Alchay, Araz Rawshani

    Aims: Acute myocardial infarction (AMI) remains a leading global cause of mortality, where timely diagnosis is critical to enable early intervention. The 12-lead electrocardiogram (ECG) is a critical tool for AMI detection. While deep learning (DL) models show promise for automated ECG analysis, most prior studies rely on small, curated datasets with limited external validation, limiting their clinical applicability. Methods and results: We developed a multimodal DL model (Conv-BiLSTM-Attn) integrating convolutional and recurrent neural networks with an attention mechanism. Using a large, multi-centre dataset of 145 656 ECGs from 96 813 patients across three Swedish hospitals. We trained the model to detect AMI using raw 12-lead ECG signals and demographic inputs (age, sex). Model performance was evaluated under two external validation protocols: generalization across hospitals (GAH) and leave-one-hospital-out (LOHO). The model achieved an area under the receiver operating characteristic (AUROC) of 0.848 (95% CI: 0.84–0.86) and an area under the precision-recall (AUPRC) of 0.456, reflecting class imbalance (∼6–10% AMI prevalence) under the GAH protocol. Subgroup AUROCs ranged from 0.79 to 0.92 across age and sex groups. At the Youden-optimized threshold (0.439), the model showed sensitivity of 0.736, specificity of 0.793, negative predictive value of 0.976, and weighted F1-score of 0.837. Under the LOHO protocol, AUROCs ranged from 0.801–0.849. At Youden thresholds (0.34–0.50), sensitivity ranged from 0.671 to 0.776 and specificity from 0.651 to 0.801, confirming generalizability across sites. Conv-BiLSTM-Attn outperformed benchmark models, with Score-CAM highlighting relevant ST–T segments. Conclusion: This DL model can support accurate and generalizable AMI detection from routine ECGs, with the Conv-BiLSTM-Attn architecture outperforming current benchmark approaches.


  • Open Access

    Fully Automated Diagnosis of Acute Myocardial Infarction Using Electrocardiograms and Multimodal Deep Learning

    by Lukas Hilgendorf, Petur Petursson, Erik Andersson, Aidin Rawshani, Deepak L. Bhatt, Truls Råmunddal, Vibha Gupta, Kristofer Skoglund, Elmir Omerovic, Helen Sjöland, et al.


  • Open Access

    End-to-end deep-learning model for the detection of coronary artery stenosis on coronary CT images

    by Vibha Gupta, Petur Petursson, Aidin Rawshani, Jan Boren, Truls Ramunddal, Deepak L Bhatt, Elmir Omerovic, Oskar Angerås, Gustav Smith, Naveed Sattar, et al.

    Purpose: We examined whether end-to-end deep-learning models could detect moderate (≥50%) or severe (≥70%) stenosis in the left anterior descending artery (LAD), right coronary artery (RCA) or left circumflex artery (LCX) in iodine contrast-enhanced ECG-gated coronary CT angiography (CCTA) scans. Methods: From a database of 6293 CCTA scans, we used pre-existing curved multiplanar reformations (CMR) images of the LAD, RCA and LCX arteries to create end-to-end deep-learning models for the detection of moderate or severe stenoses. We preprocessed the images by exploiting domain knowledge and employed a transfer learning approach using EfficientNet, ResNet, DenseNet and Inception-ResNet, with a class-weighted strategy optimised through cross-validation. Heatmaps were generated to indicate critical areas identified by the models, aiding clinicians in understanding the model’s decision-making process. Results: Among the 900 CMR cases, 279 involved the LAD artery, 259 the RCA artery and 253 the LCX artery. EfficientNet models outperformed others, with EfficientNetB3 and EfficientNetB0 demonstrating the highest accuracy for LAD, EfficientNetB2 for RCA and EfficientNetB0 for LCX. The area under the curve for receiver operating characteristic (AUROC) reached 0.95 for moderate and 0.94 for severe stenosis in the LAD. For the RCA, the AUROC was 0.92 for both moderate and severe stenosis detection. The LCX achieved an AUROC of 0.88 for the detection of moderate stenoses, though the calibration curve exhibited significant overestimation. Calibration curves matched probabilities for the LAD but showed discrepancies for the RCA. Heatmap visualisations confirmed the models’ precision in delineating stenotic lesions. Decision curve analysis and net reclassification index assessments reinforced the efficacy of EfficientNet models, confirming their superior diagnostic capabilities. Conclusion: Our end-to-end deep-learning model demonstrates, for the LAD artery, excellent discriminatory ability and calibration during internal validation, despite a small dataset used to train the network. The model reliably produces precise, highly interpretable images.


  • Open Access

    Predicting troponin biomarker elevation from electrocardiograms using a deep neural network

    by Lukas Hilgendorf, Petur Petursson, Vibha Gupta, Truls Ramunddal, Erik Andersson, Peter Lundgren, Christian Dworeck, Charlotta Ljungman, Jan Boren, Aidin Rawshani, et al.

    Background: Elevated troponin levels are a sensitive biomarker for cardiac injury. The quick and reliable prediction of troponin elevation for patients with chest pain from readily available ECGs may pose a valuable time-saving diagnostic tool during decision-making concerning this patient population. Methods and results: The data used included 15 856 ECGs from patients presenting to the emergency rooms with chest pain or dyspnoea at two centres in Sweden from 2015 to June 2023. All patients had high-sensitivity troponin test results within 6 hours after 12-lead ECG. Both troponin I (TnI) and TnT were used, with biomarker-specific cut-offs and sex-specific cut-offs for TnI. On this dataset, a residual convolutional neural network (ResNet) was trained 10 times, each on a unique split of the data. The final model achieved an average area under the curve for the receiver operating characteristic curve of 0.7717 (95% CI±0.0052), calibration curve analysis revealed a mean slope of 1.243 (95% CI±0.075) and intercept of −0.073 (95% CI±0.034), indicating a good correlation between prediction and ground truth. Post-classification, tuned for F1 score, accuracy was 71.43% (95% CI±1.28), with an F1 score of 0.5642 (95% CI±0.0052) and a negative predictive value of 0.8660 (95% CI±0.0048), respectively. The ResNet displayed comparable or surpassing metrics to prior presented models. Conclusion: The model exhibited clinically meaningful performance, notably its high negative predictive accuracy. Therefore, clinical use of comparable neural networks in first-line, quick-response triage of patients with chest pain or dyspnoea appears as a valuable option in future medical practice.