MRCON-Net: Multiscale reweighted convolutional coding neural network for low-dose CT imaging

Impact of intelligent convolutional neural network -based algorithms on head computed tomography evaluation and comprehensive rehabilitation acupuncture therapy for patients with cerebral infarction

This work was to evaluate the impacts of comprehensive rehabilitation acupuncture therapy on the recovery of neurological function in cerebral infarction (CI) patients and to utilize convolutional neural network (CNN) intelligent algorithms to optimize head computed tomography (CT) images and improve lesion localization accuracy. 98 CI patients were divided into a control group (Ctrl group) and an experimental group (Exp group), with 48 patients in each group. The patients in the Ctrl group received CT evaluation combined with comprehensive rehabilitation acupuncture therapy. While, those in the Exp group received CT evaluation with the use of CNN algorithms for optimization, along with comprehensive rehabilitation acupuncture therapy. Acupuncture therapy included selecting acupoints on the patient's head, selecting two horizontal needling needles from top to bottom at the acupoints on the front side of the lesion, and then horizontal needling along the top midline. The differences in treatment outcomes were compared between the two groups based on Fugl-Meyer upper limb assessment (FMA) scores, Barthel Index (BI) scores, National Institutes of Health Stroke Scale (NIHSS4) scores, Modified Edinburgh-Scandinavian Stroke Scale (MESSS) scores, and hemodynamics. Simultaneously, the CT images were optimized using CNN intelligent algorithms to improve image quality and lesion localization accuracy. The results showed that the CI CT images processed by the CNN-based intelligent algorithm showed significant improvements in clarity and contrast compared to conventional CT images. The CNN-based intelligent algorithm demonstrated higher sensitivity (97.5 %, 93.8 %), higher PSNR (30.14 dB, 24.72 dB), and lower missed detection rate (0.52 %, 1.88 %) in detecting CI lesions. The total effective rate in the Exp group was 95.83 %, which was significantly higher than the 85.42 % in the Ctrl group (P < 0.05). The Exp group showed significantly higher levels in FMA and BI scores (P < 0.05). After treatment, the NIHSS4 and MESSS scores in the Exp group were lower than those in the Ctrl group (P < 0.05). Additionally, post-treatment, the plasma concentrations and whole-blood viscosity (low shear and high shear) in the Exp group were lower than those in the Ctrl group, and the plasma concentration and whole-blood viscosity (high shear) were also lower than those in the Ctrl group (P < 0.05). In conclusion, comprehensive rehabilitation acupuncture therapy had a positive impact on the recovery of neurological function in CI patients. By applying CNN-based intelligent algorithms to optimize head CT images, lesion localization accuracy can be improved, thereby guiding rehabilitation treatment more effectively.

A scoping review of interpretability and explainability concerning artificial intelligence methods in medical imaging

PURPOSE

To review eXplainable Artificial Intelligence/(XAI) methods available for medical imaging/(MI).

METHOD

A scoping review was conducted following the Joanna Briggs Institute's methodology. The search was performed on Pubmed, Embase, Cinhal, Web of Science, BioRxiv, MedRxiv, and Google Scholar. Studies published in French and English after 2017 were included. Keyword combinations and descriptors related to explainability, and MI modalities were employed. Two independent reviewers screened abstracts, titles and full text, resolving differences through discussion.

RESULTS

228 studies met the criteria. XAI publications are increasing, targeting MRI (n = 73), radiography (n = 47), CT (n = 46). Lung (n = 82) and brain (n = 74) pathologies, Covid-19 (n = 48), Alzheimer's disease (n = 25), brain tumors (n = 15) are the main pathologies explained. Explanations are presented visually (n = 186), numerically (n = 67), rule-based (n = 11), textually (n = 11), and example-based (n = 6). Commonly explained tasks include classification (n = 89), prediction (n = 47), diagnosis (n = 39), detection (n = 29), segmentation (n = 13), and image quality improvement (n = 6). The most frequently provided explanations were local (78.1 %), 5.7 % were global, and 16.2 % combined both local and global approaches. Post-hoc approaches were predominantly employed. The used terminology varied, sometimes indistinctively using explainable (n = 207), interpretable (n = 187), understandable (n = 112), transparent (n = 61), reliable (n = 31), and intelligible (n = 3).

CONCLUSION

The number of XAI publications in medical imaging is increasing, primarily focusing on applying XAI techniques to MRI, CT, and radiography for classifying and predicting lung and brain pathologies. Visual and numerical output formats are predominantly used. Terminology standardisation remains a challenge, as terms like "explainable" and "interpretable" are sometimes being used indistinctively. Future XAI development should consider user needs and perspectives.

PILN: A posterior information learning network for blind reconstruction of lung CT images

BACKGROUND AND OBJECTIVE

Computer tomography (CT) imaging technology has played significant roles in the diagnosis and treatment of various lung diseases, but the degradations in CT images usually cause the loss of detailed structural information and interrupt the judgement from clinicians. Therefore, reconstructing noise-free, high-resolution CT images with sharp details from degraded ones is of great importance for the computer-assisted diagnosis (CAD) system. However, current image reconstruction methods suffer from unknown parameters of multiple degradations in actual clinical images.

METHODS

To solve these problems, we propose a unified framework, so called Posterior Information Learning Network (PILN), for blind reconstruction of lung CT images. The framework consists of two stages: Firstly, a noise level learning (NLL) network is proposed to quantify the Gaussian and artifact noise degradations into different levels. Inception-residual modules are designed to extract multi-scale deep features from the noisy image, and residual self-attention structures are proposed to refine deep features to essential representations of noise. Secondly, by taking the estimated noise levels as prior information, a cyclic collaborative super-resolution (CyCoSR) network is proposed to iteratively reconstruct the high-resolution CT image and estimate the blur kernel. Two convolutional modules are designed based on cross-attention transformer structure, named as Reconstructor and Parser. The high-resolution image is restored from the degraded image by the Reconstructor under the guidance of the predicted blur kernel, while the blur kernel is estimated by the Parser according to the reconstructed image and the degraded one. The NLL and CyCoSR networks are formulated as an end-to-end framework to handle multiple degradations simultaneously.

RESULTS

The proposed PILN is applied to the Cancer Imaging Archive (TCIA) dataset and the Lung Nodule Analysis 2016 Challenge (LUNA16) dataset to evaluate its ability in reconstructing lung CT images. Compared with the state-of-the-art image reconstruction algorithms, it can provide high-resolution images with less noise and sharper details with respect to quantitative benchmarks.

CONCLUSIONS

Extensive experimental results demonstrate that our proposed PILN can achieve better performance on blind reconstruction of lung CT images, providing noise-free, detail-sharp and high-resolution images without knowing the parameters of multiple degradation sources.