Challenges of Deep Learning in Medical Image Analysis—Improving Explainability and Trust

Dhar, Tribikram; Dey, Nilanjan; Borra, Surekha; Sherratt, R. Simon

doi:10.1109/tts.2023.3234203

Cited by 108 publications

(23 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The selected Deep Learning schema for the detection of key features of the flow waveform is most suitable for the aim of the study, since the convolutional kernels are specifically designed to identify such types of patterns that separate two classes. The shallow architecture and the use of weighted loss mitigate against the relatively small and unbalanced dataset, respectively, both of which are common issues that occur when applying deep learning algorithms to medical data [ 27 ]. Additionally, the computation method used to calculate the PTP, based on which the model was trained, was compared to experts’ manual analysis and found to be accurate.…”

Section: Discussionmentioning

confidence: 99%

Neural Network-Enabled Identification of Weak Inspiratory Efforts during Pressure Support Ventilation Using Ventilator Waveforms

Soundoulounaki

Sylligardos

Akoumianaki

et al. 2023

JPM

View full text Add to dashboard Cite

During pressure support ventilation (PSV), excessive assist results in weak inspiratory efforts and promotes diaphragm atrophy and delayed weaning. The aim of this study was to develop a classifier using a neural network to identify weak inspiratory efforts during PSV, based on the ventilator waveforms. Recordings of flow, airway, esophageal and gastric pressures from critically ill patients were used to create an annotated dataset, using data from 37 patients at 2–5 different levels of support, computing the inspiratory time and effort for every breath. The complete dataset was randomly split, and data from 22 patients (45,650 breaths) were used to develop the model. Using a One-Dimensional Convolutional Neural Network, a predictive model was developed to characterize the inspiratory effort of each breath as weak or not, using a threshold of 50 cmH2O*s/min. The following results were produced by implementing the model on data from 15 different patients (31,343 breaths). The model predicted weak inspiratory efforts with a sensitivity of 88%, specificity of 72%, positive predictive value of 40%, and negative predictive value of 96%. These results provide a ‘proof-of-concept’ for the ability of such a neural-network based predictive model to facilitate the implementation of personalized assisted ventilation.

show abstract

Section: Discussionmentioning

confidence: 99%

Neural Network-Enabled Identification of Weak Inspiratory Efforts during Pressure Support Ventilation Using Ventilator Waveforms

Soundoulounaki

Sylligardos

Akoumianaki

et al. 2023

JPM

View full text Add to dashboard Cite

show abstract

“…Although the sophisticated mathematics underlying deep learning training algorithms is conceptually understandable, their architectures are more of a "black box" paradigm. In the case of the breast density CycleGAN, one must comprehend the learned mapping via post hoc explainability [42]. The post hoc explanation is a task that will be performed in future work.…”

Section: Discussionmentioning

confidence: 99%

Breast Density Transformations Using CycleGANs for Revealing Undetected Findings in Mammograms

Anyfantis

Koutras

Apostolopoulos

et al. 2023

Signals

View full text Add to dashboard Cite

Breast cancer is the most common cancer in women, a leading cause of morbidity and mortality, and a significant health issue worldwide. According to the World Health Organization’s cancer awareness recommendations, mammographic screening should be regularly performed on middle-aged or older women to increase the chances of early cancer detection. Breast density is widely known to be related to the risk of cancer development. The American College of Radiology Breast Imaging Reporting and Data System categorizes mammography into four levels based on breast density, ranging from ACR-A (least dense) to ACR-D (most dense). Computer-aided diagnostic (CAD) systems can now detect suspicious regions in mammograms and identify abnormalities more quickly and accurately than human readers. However, their performance is still influenced by the tissue density level, which must be considered when designing such systems. In this paper, we propose a novel method that uses CycleGANs to transform suspicious regions of mammograms from ACR-B, -C, and -D levels to ACR-A level. This transformation aims to reduce the masking effect caused by thick tissue and separate cancerous regions from surrounding tissue. Our proposed system enhances the performance of conventional CNN-based classifiers significantly by focusing on regions of interest that would otherwise be misidentified due to fatty masking. Extensive testing on different types of mammograms (digital and scanned X-ray film) demonstrates the effectiveness of our system in identifying normal, benign, and malignant regions of interest.

show abstract

“…This interpretability enables doctors to make changes or enhancements to the model, ensuring its accuracy. Interpretable models also help with the ethical issues of AI use in healthcare [145]. In the context of prenatal abnormality detection, it is critical to verify that AI system decisions are fair, unbiased, and in accordance with accepted medical norms.…”

Section: Interpretable Models For Clinical Decision Supportmentioning

confidence: 99%

Ultrasound imaging based recognition of prenatal anomalies: a systematic clinical engineering review

Sriraam,

Chinta,

Suresh

et al. 2024

Prog. Biomed. Eng.

View full text Add to dashboard Cite

For prenatal screening, ultrasound imaging allows for real-time observation of developing fetal anatomy. Understanding normal and aberrant forms through extensive fetal structural assessment enables for early detection and intervention. However, the reliability of anomaly diagnosis varies depending on operator expertise and device limits. First trimester scans in conjunction with circulating biochemical markers are critical in identifying high-risk pregnancies, but they also pose technical challenges. Recent engineering advancements in automated diagnosis, such as AI-based ultrasound image processing and multimodal data fusion, are developing to improve screening efficiency, accuracy, and consistency. Still, creating trust in these data-driven solutions is necessary for integration and acceptability in clinical settings. Transparency can be promoted by explainable AI (XAI) technologies that provide visual interpretations and illustrate the underlying diagnostic decision making process. An explanatory framework based on deeplearning is suggested to construct charts depicting anomaly screening results from ultrasound video feeds. AI modeling can then be applied to these charts to connect defects with probable deformations. Overall, engineering approaches that increaseimaging, automation, and interpretability hold enormous promise for altering traditional workflows and expanding diagnostic capabilities for better prenatal care.

show abstract

Challenges of Deep Learning in Medical Image Analysis—Improving Explainability and Trust

Cited by 108 publications

References 40 publications

Neural Network-Enabled Identification of Weak Inspiratory Efforts during Pressure Support Ventilation Using Ventilator Waveforms

Neural Network-Enabled Identification of Weak Inspiratory Efforts during Pressure Support Ventilation Using Ventilator Waveforms

Breast Density Transformations Using CycleGANs for Revealing Undetected Findings in Mammograms

Ultrasound imaging based recognition of prenatal anomalies: a systematic clinical engineering review

Contact Info

Product

Resources

About