arXiv subjects

Kjersti Engan

Publications and source records attributed to Kjersti Engan.

29 records · Page 2Linked to original sources

arXiv2023

Vision Transformers for Small Histological Datasets Learned through Knowledge Distillation

Computational Pathology (CPATH) systems have the potential to automate diagnostic tasks. However, the artifacts on the digitized histological glass slides, known as Whole Slide Images (WSIs), may hamper the overall performance of CPATH systems. Deep Learning (DL) models such as Vision Transformers (ViTs) may detect and exclude artifacts before running the diagnostic algorithm. A simple way to develop robust and generalized ViTs is to train them on massive datasets. Unfortunately, acquiring large medical datasets is expensive and inconvenient, prompting the need for a generalized artifact detection method for WSIs. In this paper, we present a student-teacher recipe to improve the classification performance of ViT for the air bubbles detection task. ViT, trained under the student-teacher framework, boosts its performance by distilling existing knowledge from the high-capacity teacher model. Our best-performing ViT yields 0.961 and 0.911 F1-score and MCC, respectively, observing a 7% gain in MCC against stand-alone training. The proposed method presents a new perspective of leveraging knowledge distillation over transfer learning to encourage the use of customized transformers for efficient preprocessing pipelines in the CPATH systems.

cs.CV↗

arXiv2023

Self-supervised learning of a tailored Convolutional Auto Encoder for histopathological prostate grading

According to GLOBOCAN 2020, prostate cancer is the second most common cancer in men worldwide and the fourth most prevalent cancer overall. For pathologists, grading prostate cancer is challenging, especially when discriminating between Grade 3 (G3) and Grade 4 (G4). This paper proposes a Self-Supervised Learning (SSL) framework to classify prostate histopathological images when labeled images are scarce. In particular, a tailored Convolutional Auto Encoder (CAE) is trained to reconstruct 128x128x3 patches of prostate cancer Whole Slide Images (WSIs) as a pretext task. The downstream task of the proposed SSL paradigm is the automatic grading of histopathological patches of prostate cancer. The presented framework reports promising results on the validation set, obtaining an overall accuracy of 83% and on the test set, achieving an overall accuracy value of 76% with F1-score of 77% in G4.

eess.IV↗

arXiv2023

Self-Supervised Few-Shot Learning for Ischemic Stroke Lesion Segmentation

Precise ischemic lesion segmentation plays an essential role in improving diagnosis and treatment planning for ischemic stroke, one of the prevalent diseases with the highest mortality rate. While numerous deep neural network approaches have recently been proposed to tackle this problem, these methods require large amounts of annotated regions during training, which can be impractical in the medical domain where annotated data is scarce. As a remedy, we present a prototypical few-shot segmentation approach for ischemic lesion segmentation using only one annotated sample during training. The proposed approach leverages a novel self-supervised training mechanism that is tailored to the task of ischemic stroke lesion segmentation by exploiting color-coded parametric maps generated from Computed Tomography Perfusion scans. We illustrate the benefits of our proposed training mechanism, leading to considerable improvements in performance in the few-shot setting. Given a single annotated patient, an average Dice score of 0.58 is achieved for the segmentation of ischemic lesions.

eess.IV↗

arXiv2023

Object Detection During Newborn Resuscitation Activities

Birth asphyxia is a major newborn mortality problem in low-resource countries. International guideline provides treatment recommendations; however, the importance and effect of the different treatments are not fully explored. The available data is collected in Tanzania, during newborn resuscitation, for analysis of the resuscitation activities and the response of the newborn. An important step in the analysis is to create activity timelines of the episodes, where activities include ventilation, suction, stimulation etc. Methods: The available recordings are noisy real-world videos with large variations. We propose a two-step process in order to detect activities possibly overlapping in time. The first step is to detect and track the relevant objects, like bag-mask resuscitator, heart rate sensors etc., and the second step is to use this information to recognize the resuscitation activities. The topic of this paper is the first step, and the object detection and tracking are based on convolutional neural networks followed by post processing. Results: The performance of the object detection during activities were 96.97 % (ventilations), 100 % (attaching/removing heart rate sensor) and 75 % (suction) on a test set of 20 videos. The system also estimate the number of health care providers present with a performance of 71.16 %. Conclusion: The proposed object detection and tracking system provides promising results in noisy newborn resuscitation videos. Significance: This is the first step in a thorough analysis of newborn resuscitation episodes, which could provide important insight about the importance and effect of different newborn resuscitation activities

cs.CV↗

arXiv2023

Activity Recognition From Newborn Resuscitation Videos

Objective: Birth asphyxia is one of the leading causes of neonatal deaths. A key for survival is performing immediate and continuous quality newborn resuscitation. A dataset of recorded signals during newborn resuscitation, including videos, has been collected in Haydom, Tanzania, and the aim is to analyze the treatment and its effect on the newborn outcome. An important step is to generate timelines of relevant resuscitation activities, including ventilation, stimulation, suction, etc., during the resuscitation episodes. Methods: We propose a two-step deep neural network system, ORAA-net, utilizing low-quality video recordings of resuscitation episodes to do activity recognition during newborn resuscitation. The first step is to detect and track relevant objects using Convolutional Neural Networks (CNN) and post-processing, and the second step is to analyze the proposed activity regions from step 1 to do activity recognition using 3D CNNs. Results: The system recognized the activities newborn uncovered, stimulation, ventilation and suction with a mean precision of 77.67 %, a mean recall of 77,64 %, and a mean accuracy of 92.40 %. Moreover, the accuracy of the estimated number of Health Care Providers (HCPs) present during the resuscitation episodes was 68.32 %. Conclusion: The results indicate that the proposed CNN-based two-step ORAAnet could be used for object detection and activity recognition in noisy low-quality newborn resuscitation videos. Significance: A thorough analysis of the effect the different resuscitation activities have on the newborn outcome could potentially allow us to optimize treatment guidelines, training, debriefing, and local quality improvement in newborn resuscitation.

eess.IV↗

arXiv2023

Deep Learning for Predicting Metastasis on Melanoma WSIs

Northern Europe has the second highest mortality rate of melanoma globally. In 2020, the mortality rate of melanoma rose to 1.9 per 100 000 habitants. Melanoma prognosis is based on a pathologist's subjective visual analysis of the patient's tumor. This methodology is heavily time-consuming, and the prognosis variability among experts is notable, drastically jeopardizing its reproducibility. Thus, the need for faster and more reproducible methods arises. Machine learning has paved its way into digital pathology, but so far, most contributions are on localization, segmentation, and diagnostics, with little emphasis on prognostics. This paper presents a convolutional neural network (CNN) method based on VGG16 to predict melanoma prognosis as the presence of metastasis within five years. Patches are extracted from regions of interest from Whole Slide Images (WSIs) at different magnification levels used in model training and validation. Results infer that utilizing WSI patches at 20x magnification level has the best performance, with an F1 score of 0.7667 and an AUC of 0.81.

eess.IV↗

arXiv2023

Active Learning Based Domain Adaptation for Tissue Segmentation of Histopathological Images

Accurate segmentation of tissue in histopathological images can be very beneficial for defining regions of interest (ROI) for streamline of diagnostic and prognostic tasks. Still, adapting to different domains is essential for histopathology image analysis, as the visual characteristics of tissues can vary significantly across datasets. Yet, acquiring sufficient annotated data in the medical domain is cumbersome and time-consuming. The labeling effort can be significantly reduced by leveraging active learning, which enables the selective annotation of the most informative samples. Our proposed method allows for fine-tuning a pre-trained deep neural network using a small set of labeled data from the target domain, while also actively selecting the most informative samples to label next. We demonstrate that our approach performs with significantly fewer labeled samples compared to traditional supervised learning approaches for similar F1-scores, using barely a 59\% of the training set. We also investigate the distribution of class balance to establish annotation guidelines.

eess.IV↗

arXiv2022

Multi-input segmentation of damaged brain in acute ischemic stroke patients using slow fusion with skip connection

Time is a fundamental factor during stroke treatments. A fast, automatic approach that segments the ischemic regions helps treatment decisions. In clinical use today, a set of color-coded parametric maps generated from computed tomography perfusion (CTP) images are investigated manually to decide a treatment plan. We propose an automatic method based on a neural network using a set of parametric maps to segment the two ischemic regions (core and penumbra) in patients affected by acute ischemic stroke. Our model is based on a convolution-deconvolution bottleneck structure with multi-input and slow fusion. A loss function based on the focal Tversky index addresses the data imbalance issue. The proposed architecture demonstrates effective performance and results comparable to the ground truth annotated by neuroradiologists. A Dice coefficient of 0.81 for penumbra and 0.52 for core over the large vessel occlusion test set is achieved. The full implementation is available at: https://git.io/JtFGb.

eess.IV↗

arXiv2022

Nested Multiple Instance Learning with Attention Mechanisms

Strongly supervised learning requires detailed knowledge of truth labels at instance levels, and in many machine learning applications this is a major drawback. Multiple instance learning (MIL) is a popular weakly supervised learning method where truth labels are not available at instance level, but only at bag-of-instances level. However, sometimes the nature of the problem requires a more complex description, where a nested architecture of bag-of-bags at different levels can capture underlying relationships, like similar instances grouped together. Predicting the latent labels of instances or inner-bags might be as important as predicting the final bag-of-bags label but is lost in a straightforward nested setting. We propose a Nested Multiple Instance with Attention (NMIA) model architecture combining the concept of nesting with attention mechanisms. We show that NMIA performs as conventional MIL in simple scenarios and can grasp a complex scenario providing insights to the latent labels at different levels.

cs.LG↗

arXiv2021

CNN Based Segmentation of Infarcted Regions in Acute Cerebral Stroke Patients From Computed Tomography Perfusion Imaging

More than 13 million people suffer from ischemic cerebral stroke worldwide each year. Thrombolytic treatment can reduce brain damage but has a narrow treatment window. Computed Tomography Perfusion imaging is a commonly used primary assessment tool for stroke patients, and typically the radiologists will evaluate resulting parametric maps to estimate the affected areas, dead tissue (core), and the surrounding tissue at risk (penumbra), to decide further treatments. Different work has been reported, suggesting thresholds, and semi-automated methods, and in later years deep neural networks, for segmenting infarction areas based on the parametric maps. However, there is no consensus in terms of which thresholds to use, or how to combine the information from the parametric maps, and the presented methods all have limitations in terms of both accuracy and reproducibility. We propose a fully automated convolutional neural network based segmentation method that uses the full four-dimensional computed tomography perfusion dataset as input, rather than the pre-filtered parametric maps. The suggested network is tested on an available dataset as a proof-of-concept, with very encouraging results. Cross-validated results show averaged Dice score of 0.78 and 0.53, and an area under the receiver operating characteristic curve of 0.97 and 0.94 for penumbra and core respectively

eess.IV↗

arXiv2019

Multiscale Deep Neural Networks for Multiclass Tissue Classification of Histological Whole-Slide Images

Correct treatment of urothelial carcinoma patients is dependent on accurate grading and staging of the cancer tumour. This is determined manually by a pathologist by examining the histological whole-slide images (WSI). The large size of these images makes this a time-consuming and challenging task. The WSI contain a variety of tissue types, and a method for defining diagnostic relevant regions would have several advantages for visualization as well as further input to automated diagnosis systems. We propose an automatic multiscale method for classification of tiles from WSI of urothelial carcinoma patients into six classes. Three architectures based on convolutional neural network (CNN) were tested: MONO-CNN (400x), DI-CNN (100x/400x) and TRI-CNN (25x/100x/400x). The preliminary results show that the two multiscale models performed significantly better than the mono-scale model, achieving an F1-score of 0.986, substantiating that utilising multiple scales in the model aids the classification accuracy.

eess.IV↗