Localization of Critical Findings in Chest X-Ray Without Local Annotations Using Multi-Instance Learning

Schwab, Evan; Gooßen, André; Deshpande, Hrishikesh; Saalbach, Axel

doi:10.1109/isbi45749.2020.9098551

Cited by 15 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These tasks are considered cornerstones in most autonomously or semi-autonomously acting machines and applications (e.g., robots and self-driving cars) and indeed are critical in AIS and medical image analysis [ 32 ]. A typical vision task such as localizing and recognizing objects of interest (e.g., critical findings in chest X-ray images) [ 33 ] can be approached either using traditional vision, or advanced deep-learning-based methods ( Figure 1 ).…”

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

“…Unlike the aforementioned traditional methods, deep learning methods learn the underlying representation of the image in an end-to-end manner and without the need to handcraft these features [ 39 ]. These methods have revolutionized the application of AI across various domains, and significantly improved performance by orders of magnitude in many areas, such as in gaming and AI [ 40 ], NLP [ 41 ], health [ 42 ], medical image analysis [ 33 ], and cyber security [ 43 ], among others. As can be seen in the schematic diagram in Figure 2 (below), the key advantage of DL methods is the ability to map a set of raw pixels in an image to a particular outcome (e.g., an object’s identity, abnormality) [ 39 ], which proved to be inherently challenging tasks for traditional CV methods.…”

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

“…This is an important and core task in medical image analysis, by which the machine not only learns to classify an image as normal or abnormal, but also to localize and highlight the area in the image that exhibits the abnormality, if it exists. Examples include the processing and analysis of magnetic resonance imaging (MRI) to detect brain tumors [ 62 ], segmenting CT scan images [ 63 ], and localizing the area of interest from chest X-Ray images [ 33 ]. Such tasks were inherently challenging for conventional ML methods, which requires the careful design of heuristics.…”

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

“…However, with DL it became possible to train an end-to-end solution to solve such complex tasks. For instance, Schwab et al presented a new approach for localizing and recognizing critical findings in chest X-ray images (CHR), using a multi-instance learning method (MIL) that combines classification and detection [ 33 ].…”

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

See 3 more Smart Citations

Artificial Intelligence Surgery: How Do We Get to Autonomous Actions in Surgery?

Gumbs

Frigerio

Spolverato

et al. 2021

Sensors

View full text Add to dashboard Cite

Most surgeons are skeptical as to the feasibility of autonomous actions in surgery. Interestingly, many examples of autonomous actions already exist and have been around for years. Since the beginning of this millennium, the field of artificial intelligence (AI) has grown exponentially with the development of machine learning (ML), deep learning (DL), computer vision (CV) and natural language processing (NLP). All of these facets of AI will be fundamental to the development of more autonomous actions in surgery, unfortunately, only a limited number of surgeons have or seek expertise in this rapidly evolving field. As opposed to AI in medicine, AI surgery (AIS) involves autonomous movements. Fortuitously, as the field of robotics in surgery has improved, more surgeons are becoming interested in technology and the potential of autonomous actions in procedures such as interventional radiology, endoscopy and surgery. The lack of haptics, or the sensation of touch, has hindered the wider adoption of robotics by many surgeons; however, now that the true potential of robotics can be comprehended, the embracing of AI by the surgical community is more important than ever before. Although current complete surgical systems are mainly only examples of tele-manipulation, for surgeons to get to more autonomously functioning robots, haptics is perhaps not the most important aspect. If the goal is for robots to ultimately become more and more independent, perhaps research should not focus on the concept of haptics as it is perceived by humans, and the focus should be on haptics as it is perceived by robots/computers. This article will discuss aspects of ML, DL, CV and NLP as they pertain to the modern practice of surgery, with a focus on current AI issues and advances that will enable us to get to more autonomous actions in surgery. Ultimately, there may be a paradigm shift that needs to occur in the surgical community as more surgeons with expertise in AI may be needed to fully unlock the potential of AIS in a safe, efficacious and timely manner.

show abstract

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

Section: Deep Learning and Computer Visionmentioning

confidence: 99%

See 2 more Smart Citations

Artificial Intelligence Surgery: How Do We Get to Autonomous Actions in Surgery?

Gumbs

Frigerio

Spolverato

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Hou et al [31] have trained a CNN classifier that aggregates patch-level predictions to automatically locate discriminative patches within a whole slide tissue image, by formulating a novel Expectation-Maximization (EM) MIL based algorithm. Evan et al [32] took advantage of MIL to improve the explainability of their detection algorithm, by jointly performing classification and localization of critical findings in X-rays. In our work, in order to incorporate examples that are labelled for a classification task (i.e.…”

Section: Related Workmentioning

confidence: 99%

Learning to Localize Objects Using Limited Annotation, With Applications to Thoracic Diseases

2021

View full text Add to dashboard Cite

The localization of objects in images is a longstanding objective within the field of image processing. Most current techniques are based on machine learning approaches, which typically require careful annotation of training samples in the form of expensive bounding box labels. The need for such large-scale annotation has only been exacerbated by the widespread adoption of deep learning techniques within the image processing community: deep learning is notoriously data-hungry. Method: In this work, we attack this problem directly by providing a new method for learning to localize objects with limited annotation: most training images can simply be annotated with their whole image labels (and no bounding box), with only a small fraction marked with bounding boxes. The training is driven by a novel loss function, which is a continuous relaxation of a well-defined discrete formulation of weakly supervised learning. Care is taken to ensure that the loss is numerically well-posed. Additionally, we propose a neural network architecture which accounts for both patch dependence, through the use of Conditional Random Field layers, and shift-invariance, through the inclusion of anti-aliasing filters. Results: We demonstrate our method on the task of localizing thoracic diseases in chest X-ray images, achieving state-of-the-art performance on the ChestX-ray14 dataset. We further show that with a modicum of additional effort our technique can be extended from object localization to object detection, attaining high quality results on the Kaggle RSNA Pneumonia Detection Challenge. Conclusion: The technique presented in this paper has the potential to enable high accuracy localization in regimes in which annotated data is either scarce or expensive to acquire. Future work will focus on applying the ideas presented in this paper to the realm of semantic segmentation.

show abstract

Evaluating Local Explainable AI Techniques for the Classification of Chest X-Ray Images

Sciacca,

Estatico,

Verda

et al. 2024

Communications in Computer and Information Science

View full text Add to dashboard Cite

Localization of Critical Findings in Chest X-Ray Without Local Annotations Using Multi-Instance Learning

Cited by 15 publications

References 8 publications

Artificial Intelligence Surgery: How Do We Get to Autonomous Actions in Surgery?

Artificial Intelligence Surgery: How Do We Get to Autonomous Actions in Surgery?

Learning to Localize Objects Using Limited Annotation, With Applications to Thoracic Diseases

Evaluating Local Explainable AI Techniques for the Classification of Chest X-Ray Images

Contact Info

Product

Resources

About