Vision Transformers (ViTs) suffer from high parameter complexity because they rely on Multi-layer Perceptrons (MLPs) for nonlinear representation. This issue is particularly challenging in medical image analysis, where labeled data is limited, leading to inadequate feature representation. Existing methods have attempted to optimize either the patch embedding stage or the non-embedding stage of ViTs. Still, they have struggled to balance effective modeling, parameter complexity, and data availability. Recently, the Kolmogorov-Arnold Network (KAN) was introduced as an alternative to MLPs, offering a potential solution to the large parameter issue in ViTs. However, KAN cannot be directly integrated into ViT due to challenges such as handling 2D structured data and dimensionality catastrophe. To solve this problem, we propose MedKAFormer, the first ViT model to incorporate the Kolmogorov-Arnold (KA) theorem for medical image representation. It includes a Dynamic Kolmogorov-Arnold Convolution (DKAC) layer for flexible nonlinear modeling in the patch embedding stage. Additionally, it introduces a Nonlinear Sparse Token Mixer (NSTM) and a Nonlinear Dynamic Filter (NDF) in the non-embedding stage. These components provide comprehensive nonlinear representation while reducing model overfitting. MedKAFormer reduces parameter complexity by 85.61% compared to ViT-Base and achieves competitive results on 14 medical datasets across various imaging modalities and structures.

Cardiovascular disease is a leading global cause of death, requiring accurate heart segmentation for diagnosis and surgical planning. Deep learning methods have been demonstrated to achieve superior performances in cardiac structures segmentation. However, there are still limitations in 3D whole heart segmentation, such as inadequate spatial context modeling, difficulty in capturing long-distance dependencies, high computational complexity, and limited representation of local high-level semantic information. To tackle the above problems, we propose a lightweight Pyramid Pooling Transformer-CNN (P2TC) network for accurate 3D whole heart segmentation. The proposed architecture comprises a dual encoder-decoder structure with a 3D pyramid pooling Transformer for multi-scale information fusion and a lightweight large-kernel Convolutional Neural Network (CNN) for local feature extraction. The decoder has two branches for precise segmentation and contextual residual handling. The first branch is used to generate segmentation masks for pixel-level classification based on the features extracted by the encoder to achieve accurate segmentation of cardiac structures. The second branch highlights contextual residuals across slices, enabling the network to better handle variations and boundaries. Extensive experimental results on the Multi-Modality Whole Heart Segmentation (MM-WHS) 2017 challenge dataset demonstrate that P2TC outperforms the most advanced methods, achieving the Dice scores of 92.6% and 88.1% in Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) modalities respectively, which surpasses the baseline model by 1.5% and 1.7%, and achieves state-of-the-art segmentation results.

Nonspecific symptoms and variability in radiographic reporting patterns contribute to a diagnostic delay of the diagnosis of pulmonary fibrosis. An attractive solution is the use of machine-learning algorithms to screen for radiographic features suggestive of pulmonary fibrosis. Thus, we developed and validated a machine learning classifier algorithm (ScreenDx) to screen computed tomography imaging and identify incidental cases of pulmonary fibrosis. ScreenDx is a deep learning convolutional neural network that was developed from a multi-source dataset (cohort A) of 3,658 cases of normal and abnormal CT's, including CT's from patients with COPD, emphysema, and community-acquired pneumonia. Cohort B, a US-based cohort (n = 381) was used for tuning the algorithm, and external validation was performed on cohort C (n = 683), a separate international dataset. At the optimal threshold, the sensitivity and specificity for detection of pulmonary fibrosis in cohort B was 0.91 (95 % CI 88-94 %) and 0.95 (95 % CI 93-97 %), respectively, with AUC 0.98. In the external validation dataset (cohort C), the sensitivity and specificity were 1.0 (95 % 99.9-100.0) and 0.98 (95 % CI 97.9-99.6), respectively, with AUC 0.997. There were no significant differences in the ability of ScreenDx to identify pulmonary fibrosis based on CT manufacturer (Phillips, Toshiba, GE Healthcare, or Siemens) or slice thickness (2 mm vs 2-4 mm vs 4 mm). Regardless of CT manufacturer or slice thickness, ScreenDx demonstrated high performance across two, multi-site datasets for identifying incidental cases of pulmonary fibrosis. This suggest that the algorithm may be generalizable across patient populations and different healthcare systems.

Identifying patients likely to regain consciousness early on is a challenge. The assessment of consciousness levels and the prediction of wakefulness probabilities are facilitated by ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography (PET). This study aimed to develop a prognostic model for predicting 1-year postinjury outcomes in prolonged disorders of consciousness (DoC) using ¹⁸F-FDG PET alongside clinical behavioral scores. Eighty-seven patients with prolonged DoC newly diagnosed with behavioral Coma Recovery Scale-Revised (CRS-R) scores and ¹⁸F-FDG PET/computed tomography (18F-FDG PET/CT) scans were included. PET images were normalized by the cerebellum and extracerebral tissue, respectively. Images were divided into training and independent test sets at a ratio of 5:1. Image-based classification was conducted using the DenseNet121 network, whereas tabular-based deep learning was employed to train depth features extracted from imaging models and behavioral CRS-R scores. The performance of the models was assessed and compared using the McNemar test. Among the 87 patients with DoC who received routine treatments, 52 patients showed recovery of consciousness, whereas 35 did not. The classification of the standardized uptake value ratio by extracerebral tissue model demonstrated a higher specificity and lower sensitivity in predicting consciousness recovery than the classification of the standardized uptake value ratio by cerebellum model. With area under the curve values of 0.751 ± 0.093 and 0.412 ± 0.104 on the test sets, respectively, the difference is not statistically significant (P = 0.73). The combination of standardized uptake value ratio by extracerebral tissue and computed tomography depth features with behavioral CRS-R scores yielded the highest classification accuracy, with area under the curve values of 0.950 ± 0.027 and 0.933 ± 0.015 on the training and test sets, respectively, outperforming any individual mode. In this preliminary study, a multimodal prognostic model based on ¹⁸F-FDG PET extracerebral normalization and behavioral CRS-R scores facilitated the prediction of recovery in DoC.

Motion artifacts compromise the quality of magnetic resonance imaging (MRI) and pose challenges to achieving diagnostic outcomes and image-guided therapies. In recent years, supervised deep learning approaches have emerged as successful solutions for motion artifact reduction (MAR). One disadvantage of these methods is their dependency on acquiring paired sets of motion artifact-corrupted (MA-corrupted) and motion artifact-free (MA-free) MR images for training purposes. Obtaining such image pairs is difficult and therefore limits the application of supervised training. In this paper, we propose a novel UNsupervised Abnormality Extraction Network (UNAEN) to alleviate this problem. Our network is capable of working with unpaired MA-corrupted and MA-free images. It converts the MA-corrupted images to MA-reduced images by extracting abnormalities from the MA-corrupted images using a proposed artifact extractor, which intercepts the residual artifact maps from the MA-corrupted MR images explicitly, and a reconstructor to restore the original input from the MA-reduced images. The performance of UNAEN was assessed by experimenting with various publicly available MRI datasets and comparing them with state-of-the-art methods. The quantitative evaluation demonstrates the superiority of UNAEN over alternative MAR methods and visually exhibits fewer residual artifacts. Our results substantiate the potential of UNAEN as a promising solution applicable in real-world clinical environments, with the capability to enhance diagnostic accuracy and facilitate image-guided therapies.

Cervical length (CL) measurement using transvaginal ultrasound is an effective screening tool to assess the risk of preterm birth. An adequate assessment of CL is crucial, however, manual sonographic CL measurement is highly operator-dependent and cumbersome. Therefore, a reliable and reproducible automatic method for CL measurement is in high demand to reduce inter-rater variability and improve workflow. Despite the increasing use of artificial intelligence techniques in ultrasound, applying deep learning (DL) to analyze ultrasound images of the cervix remains a challenge due to low signal-to-noise ratios and difficulties in capturing the cervical canal, which appears as a thin line and with extremely low contrast against the surrounding tissues. To address these challenges, we have developed CL-Net, a novel DL network that incorporates expert anatomical knowledge to identify the cervix, similar to the approach taken by clinicians. CL-Net captures anatomical features related to CL measurement, facilitating the identification of the cervical canal. It then identifies the cervical canal and automatically provides reproducible and reliable CL measurements. CL-Net achieved a success rate of 95.5% in recognizing the cervical canal, comparable to that of human experts (96.4%). Furthermore, the differences between the CL measurements of CL-Net and ground truth were considerably smaller than those made by non-experts and were comparable to those made by experts (median 1.36 mm, IQR 0.87-2.82 mm, range 0.06-6.95 mm for straight cervix; median 1.31 mm, IQR 0.61-2.65 mm, range 0.01-8.18 mm for curved one).

Breast cancer is a devastating disease that affects women worldwide, and computer-aided algorithms have shown potential in automating cancer diagnosis. Recently Generative Artificial Intelligence (GenAI) opens new possibilities for addressing the challenges of labeled data scarcity and accurate prediction in critical applications. However, a lack of diversity, as well as unrealistic and unreliable data, have a detrimental impact on performance. Therefore, this study proposes an augmentation scheme to address the scarcity of labeled data and data imbalance in medical datasets. This approach integrates the concepts of the Gaussian-Laplacian pyramid and pyramid blending with similarity measures. In order to maintain the structural properties of images and capture inter-variability of patient images of the same category similarity-metric-based intermixing has been introduced. It helps to maintain the overall quality and integrity of the dataset. Subsequently, deep learning approach with significant modification, that leverages transfer learning through the usage of concatenated pre-trained models is applied to classify breast cancer histopathological images. The effectiveness of the proposal, including the impact of data augmentation, is demonstrated through a detailed analysis of three different medical datasets, showing significant performance improvement over baseline models. The proposal has the potential to contribute to the development of more accurate and reliable approach for breast cancer diagnosis.

The size of an individual's tongue relative to the oral cavity is associated with articulation speed [Feng, Lu, Zheng, Chi, and Honda, in Proceedings of the 10th Biennial Asia Pacific Conference on Speech, Language, and Hearing (2017), pp. 17-19)] and may affect speech clarity. This study introduces an ultrasound-based method for measuring relative tongue size, termed ultrasound-based relative tongue size (uRTS), as a cost-effective alternative to the magnetic resonance imaging (MRI) based method. Using deep learning to extract the tongue contour, uRTS was calculated from tongue and oropharyngeal cavity sizes in the midsagittal plane. Results from ten speakers showed a strong correlation between uRTS and MRI-based measurements (r = 0.87) and a negative correlation with tongue movement speed (r = -0.73), indicating uRTS is a useful index for assessing tongue size.

Thyroid nodules classified within the Thyroid Imaging Reporting and Data Systems (TI-RADS) category 3-5 are typically regarded as having varying degrees of malignancy risk, with the risk increasing from TI-RADS 3 to TI-RADS 5. While some of these nodules may undergo fine-needle aspiration (FNA) biopsy to assess their nature, this procedure carries a risk of false negatives and inherent complications. To avoid the need for unnecessary biopsy examination, we explored a method for distinguishing the benign and malignant characteristics of thyroid TI-RADS 3-5 nodules based on deep-learning ultrasound images combined with computed tomography (CT). Thyroid nodules, assessed as American College of Radiology (ACR) TI-RADS category 3-5 through conventional ultrasound, all of which had postoperative pathology results, were examined using both conventional ultrasound and CT before operation. We investigated the effectiveness of deep-learning models based on ultrasound alone, CT alone, and a combination of both imaging modalities using the following metrics: Area Under Curve (AUC), sensitivity, accuracy, and positive predictive value (PPV). Additionally, we compared the diagnostic efficacy of the combined methods with manual readings of ultrasound and CT. A total of 768 thyroid nodules falling within TI-RADS categories 3-5 were identified across 768 patients. The dataset comprised 499 malignant and 269 benign cases. For the automatic identification of thyroid TI-RADS category 3-5 nodules, deep learning combined with ultrasound and CT demonstrated a significantly higher AUC (0.930; 95% CI: 0.892, 0.969) compared to the application of ultrasound alone AUC (0.901; 95% CI: 0.856, 0.947) or CT alone AUC (0.776; 95% CI: 0.713, 0.840). Additionally, the AUC of combined modalities surpassed that of radiologists'assessments using ultrasound alone AUCmean (0.725；95% CI:0.677, 0.773), CT alone AUCmean (0.617; 95% CI:0.564, 0.669). Deep learning method combined with ultrasound and CT imaging of thyroid can allow more accurate and precise classification of nodules within TI-RADS categories 3-5.

Integrating artificial intelligence (AI) in the dental field can potentially enhance the efficiency of dental care. However, few studies have investigated whether AI software can achieve results comparable to those obtained by dental practitioners (general practitioners (GPs) and specialists) when assessing alveolar bone loss in a clinical setting. Thus, this study compared the performance of AI in assessing periodontal bone loss with those of GPs and specialists. This comparative cross-sectional study evaluated the performance of dental practitioners and AI software in assessing alveolar bone loss. Radiographs were randomly selected to ensure representative samples. Dental practitioners independently evaluated the radiographs, and the AI software "Second Opinion Software" was tested using the same set of radiographs evaluated by the dental practitioners. The results produced by the AI software were then compared with the baseline values to measure their accuracy and allow direct comparison with the performance of human specialists. The survey received 149 responses, where each answered 10 questions to compare the measurements made by AI and dental practitioners when assessing the amount of bone loss radiographically. The mean estimates of the participants had a moderate positive correlation with the radiographic measurements (rho = 0.547, <i>p</i> < 0.001) and a weaker but still significant correlation with AI measurements (rho = 0.365, <i>p</i> < 0.001). AI measurements had a stronger positive correlation with the radiographic measurements (rho = 0.712, <i>p</i> < 0.001) compared with their correlation with the estimates of dental practitioners. This study highlights the capacity of AI software to enhance the accuracy and efficiency of radiograph-based evaluations of alveolar bone loss. Dental practitioners are vital for the clinical experience but AI technology provides a consistent and replicable methodology. Future collaborations between AI experts, researchers, and practitioners could potentially optimize patient care.