Juana González‐Bueno Puyal

Olympus (Japan)

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Peter Mountney

Olympus (Japan)

Laurence Lovat

University College London

Danail Stoyanov

Wellcome / EPSRC Centre for Interventional and Surgical Sciences

Rawen Kader

University College London

Omer F. Ahmad

Wellcome / EPSRC Centre for Interventional and Surgical Sciences

Dániel Tóth

University Hospital of Zurich

Mohamed Hussein

University College London

Patrick Brandão

Wellcome / EPSRC Centre for Interventional and Surgical Sciences

Rehan Haidry

University College Hospital

Vinay Sehgal

University College London

Cooperative Institutions

University College London

University College Hospital

University College London Hospitals NHS Foundation Trust

Wellcome / EPSRC Centre for Interventional and Surgical Sciences

Royal London Hospital

RELX Group (Netherlands)

London Vision Clinic

Cleveland Clinic London

The London College

Hudson Institute

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Spatio-temporal classification for polyp diagnosis

Biomedical Optics Express (2022)

Juana González‐Bueno Puyal Patrick Brandão Omer F. Ahmad Kanwal K. Bhatia Dániel Tóth

Colonoscopy remains the gold standard investigation for colorectal cancer screening as it offers the opportunity to both detect and resect pre-cancerous polyps. Computer-aided polyp characterisation can determine which polyps need polypectomy and recent deep learning-based approaches have shown promising results as clinical decision support tools. Yet polyp appearance during a procedure can vary, making automatic predictions unstable. In this paper, we investigate the use of spatio-temporal information to improve the performance of lesions classification as adenoma or non-adenoma. Two methods are implemented showing an increase in performance and robustness during extensive experiments both on internal and openly available benchmark datasets.

Polypectomy

Robustness

Benchmark (surveying)

Gold standard (test)

10.1364/boe.473446

Cite

Citations (1)

A new artificial intelligence system successfully detects and localises early neoplasia in Barrett's esophagus by using convolutional neural networks

United European Gastroenterology Journal (2022)

Mohamed Hussein Juana González‐Bueno Puyal David Lines Vinay Sehgal Dániel Tóth

Abstract Background and aims Seattle protocol biopsies for Barrett's Esophagus (BE) surveillance are labour intensive with low compliance. Dysplasia detection rates vary, leading to missed lesions. This can potentially be offset with computer aided detection. We have developed convolutional neural networks (CNNs) to identify areas of dysplasia and where to target biopsy. Methods 119 Videos were collected in high‐definition white light and optical chromoendoscopy with i‐scan (Pentax Hoya, Japan) imaging in patients with dysplastic and non‐dysplastic BE (NDBE). We trained an indirectly supervised CNN to classify images as dysplastic/non‐dysplastic using whole video annotations to minimise selection bias and maximise accuracy. The CNN was trained using 148,936 video frames (31 dysplastic patients, 31 NDBE, two normal esophagus), validated on 25,161 images from 11 patient videos and tested on 264 iscan‐1 images from 28 dysplastic and 16 NDBE patients which included expert delineations. To localise targeted biopsies/delineations, a second directly supervised CNN was generated based on expert delineations of 94 dysplastic images from 30 patients. This was tested on 86 i‐scan one images from 28 dysplastic patients. Findings The indirectly supervised CNN achieved a per image sensitivity in the test set of 91%, specificity 79%, area under receiver operator curve of 93% to detect dysplasia. Per‐lesion sensitivity was 100%. Mean assessment speed was 48 frames per second (fps). 97% of targeted biopsy predictions matched expert and histological assessment at 56 fps. The artificial intelligence system performed better than six endoscopists. Interpretation Our CNNs classify and localise dysplastic Barrett's Esophagus potentially supporting endoscopists during surveillance.

Barrett's esophagus

Chromoendoscopy

10.1002/ueg2.12233

Cite

Citations (33)

Polyp characterization using deep learning and a publicly accessible polyp video database

Digestive Endoscopy (2022)

Rawen Kader Anton Cid‐Mejias Patrick Brandão Shahraz Islam Sanjith Hebbar

Convolutional neural networks (CNN) for computer-aided diagnosis of polyps are often trained using high-quality still images in a single chromoendoscopy imaging modality with sessile serrated lesions (SSLs) often excluded. This study developed a CNN from videos to classify polyps as adenomatous or nonadenomatous using standard narrow-band imaging (NBI) and NBI-near focus (NBI-NF) and created a publicly accessible polyp video database.

Chromoendoscopy

Hyperplastic Polyp

Narrow-band imaging

Diminutive

Benchmarking

10.1111/den.14500

Cite

Citations (10)

Detección y Clasificación de Tejidos Anómalos en Mamografías Digitales Mediante Redes Neuronales Convolucionales.

Juana González‐Bueno Puyal Belén Ríos Sánchez Maria del Carmen Sánchez Avila

El cancer de mama constituye un problema de salud global que supone mas del 25% de los nuevos casos de cancer en mujeres y en el que la deteccion precoz mediante la realizacion de mamografias juega un papel fundamental. Este trabajo presenta un sistema novedoso de deteccion y clasificacion de anomalias en imagenes mamograficas mediante redes neuronales convolucionales (CNN). Se trata de un sistema ambicioso que permite distinguir entre cinco clases de mamografias: sin anomalias, con masas tumorales benignas, con masas tumorales malignas, con microcalcificaciones benignas o con microcalcificaciones malignas. Este trabajo evalua no solo la precision de las CNN aplicadas a este problema concreto, sino tambien la influencia de otros parametros como la inclusion de una etapa de mejora de la calidad de la imagen o la resolucion y cantidad de imagenes utilizadas para entrenar la red.

Source

Cite

Citations (0)

Polyp detection on video colonoscopy using a hybrid 2D/3D CNN

Medical Image Analysis (2022)

Juana González‐Bueno Puyal Patrick Brandão Omer F. Ahmad Kanwal K. Bhatia Dániel Tóth

Colonoscopy is the gold standard for early diagnosis and pre-emptive treatment of colorectal cancer by detecting and removing colonic polyps. Deep learning approaches to polyp detection have shown potential for enhancing polyp detection rates. However, the majority of these systems are developed and evaluated on static images from colonoscopies, whilst in clinical practice the treatment is performed on a real-time video feed. Non-curated video data remains a challenge, as it contains low-quality frames when compared to still, selected images often obtained from diagnostic records. Nevertheless, it also embeds temporal information that can be exploited to increase predictions stability. A hybrid 2D/3D convolutional neural network architecture for polyp segmentation is presented in this paper. The network is used to improve polyp detection by encompassing spatial and temporal correlation of the predictions while preserving real-time detections. Extensive experiments show that the hybrid method outperforms a 2D baseline. The proposed architecture is validated on videos from 46 patients and on the publicly available SUN polyp database. A higher performance and increased generalisability indicate that real-world clinical implementations of automated polyp detection can benefit from the hybrid algorithm and the inclusion of temporal information.

Implementation

Gold standard (test)

10.1016/j.media.2022.102625

Cite

Citations (26)

ID: 3522438 DEEP NEURAL NETWORK FOR THE LOCALISATION OF EARLY NEOPLASIA IN BARRETT'S ESOPHAGUS WITH TARGETED BIOPSIES

Gastrointestinal Endoscopy (2021)

Mohamed Hussein Juana González‐Bueno Puyal David Lines Vinay Sehgal Dániel Tóth

Barrett's esophagus

10.1016/j.gie.2021.03.437

Cite

Citations (0)

Endoscopic Polyp Segmentation Using a Hybrid 2D/3D CNN

Lecture notes in computer science (2020)

Juana González‐Bueno Puyal Kanwal K. Bhatia Patrick Brandão Omer F. Ahmad Dániel Tóth

Implementation

10.1007/978-3-030-59725-2_29

Cite

Citations (26)

HFR-2 Computer aided diagnosis for the characterisation of dysplasia in Barrett’s oesophagus with magnification endoscopy

Mohamed Hussein David Lines Juana González‐Bueno Puyal Nicola Bowman Vinay Sehgal

Introduction

There have been significant advances in magnification endoscopic imaging of Barrett's oesophagus (BE). Magnification endoscopy of mucosal and vascular patterns arising in BE can help predict non-dysplastic from dysplastic mucosa. This can inform sampling and guide endoscopic eradication therapy. We aimed to develop a computer aided detection system that can support the diagnosis of BE dysplasia on magnification endoscopy.

Methods

Videos were collected in high-definition magnification white light and virtual chromoendoscopy with i-scan (Pentax Hoya, Japan) imaging modes in patients with dysplastic lesions in BE (high grade dysplasia (HGD)/intramucosal adenocarcinoma) and patients with non-dysplastic BE (NDBE). Endoscopic resection margins/targeted biopsy site histology served as the ground truth for dysplasia in videos. Videos were annotated for definite visual presence of dysplasia. We trained a convolutional neural network with a Resnet101 architecture to classify video frames into dysplastic or non-dysplastic using randomly selected frames from annotated videos.

Results

58 patients each with high quality video frames of magnification areas of BE (34 dysplasia, 24 NDBE) were included. Performance was evaluated using a 15-fold cross validation methodology. 76,496 (47,438 dysplasia, 29,058 NDBE) magnification video frames were analysed by the neural network. All dysplastic and non-dysplastic frames were included. We used an exponentially weighted moving average of consecutive frames to make a diagnosis of dysplasia. The network achieved a per frame sensitivity of 82%, specificity of 82%, and Area under the ROC of 90%. The mean assessment speed per frame was 0.0135 seconds (standard deviation, + 0.006) (Figure 1).

Conclusion

The neural network can characterise BE dysplasia with high accuracy and speed on magnification endoscopic images. Whole video frames were used to train and test the data moving it towards real time automated diagnosis. This will potentially aid endoscopists to make key decisions regarding endoscopic sampling and resection in BE during the same endoscopic session.

Chromoendoscopy

Barrett's esophagus

Endoscopic mucosal resection

10.1136/gutjnl-2021-bsg.61

Cite

Citations (0)

P143 Performance of a convolutional neural network for Barrett’s oesophagus related dysplasia detection following quality labelling

Poster presentations (2022)

Nasar Aslam Mohamed Hussein Rawen Kader Anton Mejias Patrick Brandão

Introduction

Invasive oesophageal adenocarcinoma arising in Barrett's oesophagus (BE) can occur in up to 25% of patients one year after index endoscopic examination. Computer assisted diagnosis may reduce the incidence of overlooked early neoplasia in BE patients. Our group has previously trained a convolutional neural network (CNN) that achieves dysplasia detection rates comparable to expert endoscopists. The aim of this study was to evaluate the effect that image quality has on the performance of our CNN.

Methods

We previously trained a CNN with Resnet101 architecture to classify videos frames according to the presence or absence of dysplasia. All available video frames demonstrating dysplasia from high-definition white light and i-scan Pentax (Hoya, Japan) imaging modes were used as part of the training for the CNN. Histology from mucosal resection specimens or biopsies served as the ground truth for the presence of dysplasia. Low- or high-quality markers were assigned to each frame with dysplasia present. Quality labelling was assigned by a single reviewer. Factors that influenced quality labelling included overall visibility of the dysplastic lesion, motion blur, light speculation, or artefact.

Results

A total of 23513 frames of dysplastic BE were assigned quality labels across 17 cases. 17915 frames were labelled as high quality with 5598 being of low quality. Evaluating the CNN performance solely on high quality frames yielded an overall sensitivity of 84.61%. When testing against both high- and low-quality frames the CNN achieved an overall sensitivity of 82.26% (table 1).

Conclusions

Evaluating the performance of our CNN on a combination of high- and low-quality dysplastic frames yields comparable overall sensitivity to testing solely on high quality dysplastic frames. Real time endoscopy captures a mix of high- and low-quality frames which is influenced by factors such as artefact, patient tolerance and scope stability. The comparable performance of our CNN will be valuable for future real time application.

10.1136/gutjnl-2022-bsg.198

Cite

Citations (0)

Automated colonoscopy withdrawal phase duration estimation using cecum detection and surgical tasks classification

Biomedical Optics Express (2023)

Thomas De Carvalho Rawen Kader Patrick Brandão Juana González‐Bueno Puyal Laurence Lovat

Colorectal cancer is the third most common type of cancer with almost two million new cases worldwide. They develop from neoplastic polyps, most commonly adenomas, which can be removed during colonoscopy to prevent colorectal cancer from occurring. Unfortunately, up to a quarter of polyps are missed during colonoscopies. Studies have shown that polyp detection during a procedure correlates with the time spent searching for polyps, called the withdrawal time. The different phases of the procedure (cleaning, therapeutic, and exploration phases) make it difficult to precisely measure the withdrawal time, which should only include the exploration phase. Separating this from the other phases requires manual time measurement during the procedure which is rarely performed. In this study, we propose a method to automatically detect the cecum, which is the start of the withdrawal phase, and to classify the different phases of the colonoscopy, which allows precise estimation of the final withdrawal time. This is achieved using a Resnet for both detection and classification trained with two public datasets and a private dataset composed of 96 full procedures. Out of 19 testing procedures, 18 have their withdrawal time correctly estimated, with a mean error of 5.52 seconds per minute per procedure.

Withdrawal time

Insertion time

10.1364/boe.485069

Cite

Citations (1)