biostudies-literatureUnknown1(1)Ziemer PGP<h4>Aims</h4>Assessment of minimum lumen areas in intravascular ultrasound (IVUS) pullbacks is time-consuming and demands adequately trained personnel. In this work, we introduce a novel and fully automated pipeline to segment the lumen boundary in IVUS datasets.<h4>Methods and results</h4>First, an automated gating is applied to select end-diastolic frames and bypass saw-tooth artefacts. Second, within a machine learning (ML) environment, we automatically segment the lumen boundary using a multi-frame (MF) convolutional neural network (MFCNN). Finally, we use the theory of Gaussian processes (GPs) to regress the final lumen boundary. The dataset consisted of 85 IVUS pullbacks (52 patients). The dataset was partitioned at the pullback-level using 73 pullbacks for training (20 586 frames), European heart journal. Digital health75-82https://www.ebi.ac.uk/biostudies/studies/S-EPMC9707866biostudies-literatureAutomated lumen segmentation using multi-frame convolutional neural networks in intravascular ultrasound datasets.PMC9707866Garcia-Garcia HMGuedes Bezerra CLemos PABulant CAAlvarez LAMOrlando JIBlanco PJZiemer PGPMaso Talou GDfalseAutomated lumen segmentation using multi-frame convolutional neural networks in intravascular ultrasound datasets.<h4>Aims</h4>Assessment of minimum lumen areas in intravascular ultrasound (IVUS) pullbacks is time-consuming and demands adequately trained personnel. In this work, we introduce a novel and fully automated pipeline to segment the lumen boundary in IVUS datasets.<h4>Methods and results</h4>First, an automated gating is applied to select end-diastolic frames and bypass saw-tooth artefacts. Second, within a machine learning (ML) environment, we automatically segment the lumen boundary using a multi-frame (MF) convolutional neural network (MFCNN). Finally, we use the theory of Gaussian processes (GPs) to regress the final lumen boundary. The dataset consisted of 85 IVUS pullbacks (52 patients). The dataset was partitioned at the pullback-level using 73 pullbacks for training (20 586 frames), 2020-01-01T00:00:00Z2020 Nov2025-04-26T13:23:38.072Z2025-04-06T14:10:44.965ZS-EPMC97078663671396110.1093/ehjdh/ztaa014