Nina Montaña-Brown

Nina Montaña-Brown

Auto-generated profile, based on publicly available information

Nina Montaña-Brown

UCL Hawkes Institute, and Department of Medical Physics & Biomedical Engineering, University College London, London, UK

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 November 2020 in Journal of Open Source Software
Verified
Journal Article
Open access
PDF file
YFYunguan Fu
Yunguan Fu
NBNina Montaña Brown
Nina Montaña Brown
SSShaheer U. Saeed
Shaheer U. Saeed
ACAdrià Casamitjana
Adrià Casamitjana
ZBZachary M. C. Baum
Zachary M. C. Baum
YHYipeng Hu
Yipeng Hu

27 May 2021 in International Journal of Computer Assisted Radiology and Surgery
Verified
Journal Article
Open access
PDF file
NMNina Montaña-Brown
Nina Montaña-Brown
JRJoão Ramalhinho
João Ramalhinho
MAMoustafa Allam
Moustafa Allam
BDBrian Davidson
Brian Davidson
YHYipeng Hu
Yipeng Hu
MCMatthew J. Clarkson
Matthew J. Clarkson
Purpose: Registration of Laparoscopic Ultrasound (LUS) to a pre-operative scan such as Computed Tomography (CT) using blood vessel information has been proposed as a method to enable image-guidance for laparoscopic liver resection. Currently, there are solutions for this problem that can potentially enable clinical translation by bypassing the need for a manual initialisation and tracking information. However, no reliable framework for the segmentation of vessels in 2D untracked LUS images has been presented. Methods: We propose the use of 2D UNet for the segmentation of liver vessels in 2D LUS images. We integrate these results in a previously developed registration method, and show the feasibility of a fully automatic initialisation to the LUS to CT registration problem without a tracking device. Results: We validate our segmentation using LUS data from 6 patients. We test multiple models by placing patient datasets into different combinations of training, testing and hold-out, and obtain mean Dice scores ranging from 0.543 to 0.706. Using these segmentations, we obtain registration accuracies between 6.3 and 16.6 mm in 50% of cases. Conclusions: We demonstrate the first instance of deep learning (DL) for the segmentation of liver vessels in LUS. Our results show the feasibility of UNet in detecting multiple vessel instances in 2D LUS images, and potentially automating a LUS to CT registration pipeline.

13 September 2024 in medRxiv
Verified
Preprint
Open access
PDF file
EAElham Asgari
Elham Asgari
SKSaleh Khalil
Saleh Khalil
NMNina Montaña-Brown
Nina Montaña-Brown
MDMagda Dubois
Magda Dubois
JBJasmine Balloch
Jasmine Balloch
DPDominic Pimenta
Dominic Pimenta
Integrating large language models (LLMs) into healthcare settings can improve workflow efficiency and patient care by automating tasks such as summarising consultations. However, ensuring the fidelity between LLM outputs and ground truth information is crucial, as errors can lead to miscommunication between patients and clinicians, resulting in incorrect diagnoses, treatment decisions and compromised patient safety. We introduce a clinician-in-the-loop framework with: 1) a clinically and technically-informed error taxonomy to classify LLM outputs, 2) an experimental structure to comprehensively and iteratively compare outputs within our LLM document generation pipeline, 3) a clinical safety framework to assess potential harms of errors in LLM outputs, and 4) an encompassing graphical user interface (GUI), CREOLA, to perform and assess all previous steps. Our clinical error metrics were derived from 18 experimental configurations involving LLMs for clinical note generation consisting of 49,590 transcript and 12,999 clinical note sentences. Overall, we observed a 1.47% hallucination rate (44% rated ‘major’) and a 3.45% omission rate (17% ‘major’). Through iterative prompts and workflow refinements, we reduced major errors below previously reported human note-taking error rates, underscoring the potential of our framework to enable safer clinical documentation.

16 May 2023 in Journal of Open Source Software
Verified
Journal Article
Open access
PDF file
AAAna Lawry Aguila
Ana Lawry Aguila
AJAlejandra Jayme
Alejandra Jayme
NMNina Montaña-Brown
Nina Montaña-Brown
VHVincent Heuveline
Vincent Heuveline
AAAndre Altmann
Andre Altmann

16 April 2024 in Medical Image Analysis
Verified
Journal Article
Open access
SSShaheer U. Saeed
Shaheer U. Saeed
JRJoão Ramalhinho
João Ramalhinho
MPMark Pinnock
Mark Pinnock
ZSZiyi Shen
Ziyi Shen
YFYunguan Fu
Yunguan Fu
YHYipeng Hu
Yipeng Hu
Supervised machine learning-based medical image computing applications necessitate expert label curation, while unlabelled image data might be relatively abundant. Active learning methods aim to prioritise a subset of available image data for expert annotation, for label-efficient model training. We develop a controller neural network that measures priority of images in a sequence of batches, as in batch-mode active learning, for multi-class segmentation tasks. The controller is optimised by rewarding positive task-specific performance gain, within a Markov decision process (MDP) environment that also optimises the task predictor. In this work, the task predictor is a segmentation network. A meta-reinforcement learning algorithm is proposed with multiple MDPs, such that the pre-trained controller can be adapted to a new MDP that contains data from different institutes and/or requires segmentation of different organs or structures within the abdomen. We present experimental results using multiple CT datasets from more than one thousand patients, with segmentation tasks of nine different abdominal organs, to demonstrate the efficacy of the learnt prioritisation controller function and its cross-institute and cross-organ adaptability. We show that the proposed adaptable prioritisation metric yields converging segmentation accuracy for a new kidney segmentation task, unseen in training, using between approximately 40% to 60% of labels otherwise required with other heuristic or random prioritisation metrics. For clinical datasets of limited size, the proposed adaptable prioritisation offers a performance improvement of 22.6% and 10.2% in Dice score, for tasks of kidney and liver vessel segmentation, respectively, compared to random prioritisation and alternative active sampling strategies.

27 February 2019
Verified
Conference Paper
Closed access
PDF file
Wenfeng XiaWenfeng Xia
Wenfeng Xia
EMEfthymios Maneas
Efthymios Maneas
NHNam Trung Huynh
Nam Trung Huynh
Mithun Kuniyil Ajith SinghMithun Kuniyil Ajith Singh
Mithun Kuniyil Ajith Singh
NBNina Montaña Brown
Nina Montaña Brown
ADAdrien E. Desjardins
Adrien E. Desjardins
Non-invasive imaging plays an important role in diagnosing and monitoring peripheral artery disease (PAD). Doppler ultrasound imaging can be used for measuring blood flow in this context. However, this technique frequently provides low contrast for flow in small vessels. Photoacoustic imaging can allow for the visualization of blood in small vessels, with direct contrast from optical absorption of hemoglobin. In this work, we investigate the potential applications of a compact LED-based photoacoustic (850 nm) and ultrasound imaging system for visualizing human peripheral blood vessels during cuff occlusion. Each measurement comprised three stages. First, a baseline measurement of a digital artery of a human finger from a volunteer without a diagnosis of PAD was performed for several seconds. Second, arterial blood flow was stopped using an occlusion cuff, with a rapid increase of pressure up to 220 mm Hg. Third, the occlusion cuff was released rapidly. Raw photoacoustic and ultrasound image data (frame rate: 70 Hz) were recorded for the entire duration of the measurement (20 s). The average photoacoustic image amplitude over an image region that enclosed the digital artery was calculated. With this value, pulsations of image amplitudes from the arteries was clearly visualized. The average photoacoustic image amplitude decreased during the increase in cuff pressure and it was followed by a rapid recovery during cuff release. With real-time non-invasive measurements of peripheral blood vessel dynamics in vivo, the compact LED-based system could be valuable for point-of-care imaging to guide treatment of PAD.