BACKGROUND: There is little literature describing the artificial intelligence (AI)-aided diagnosis of severe pneumonia (SP) subphenotypes and the association of the subphenotypes with the ventilatory treatment efficacy. The aim of our study is to illustrate whether clinical and biological heterogeneity, such as ventilation and gas-exchange, exists among patients with SP using chest computed tomography (CT)-based AI-aided latent class analysis (LCA). METHODS: This retrospective study included 413 patients hospitalized at Xinhua Hospital diagnosed with SP from June 1, 2015 to May 30, 2020. AI quantification results of chest CT and their combination with additional clinical variables were used to develop LCA models in an SP population. The optimal subphenotypes were determined though evaluating statistical indicators of all the LCA models, and clinical implications of them such as guiding ventilation strategies were further explored by statistical methods. RESULTS: The two-class LCA model based on AI quantification results of chest CT can describe the biological characteristics of the SP population well and hence yielded the two clinical subphenotypes. Patients with subphenotype-1 had milder infections ( P <0.001) than patients with subphenotype-2 and had lower 30-day ( P <0.001) and 90-day ( P <0.001) mortality, and lower in-hospital ( P = 0.001) and 2-year ( P <0.001) mortality. Patients with subphenotype-1 showed a better match between the percentage of non-infected lung volume (used to quantify ventilation) and oxygen saturation (used to reflect gas exchange), compared with patients with subphenotype-2. There were significant differences in the matching degree of lung ventilation and gas exchange between the two subphenotypes ( P <0.001). Compared with patients with subphenotype-2, those with subphenotype-1 showed a relatively better match between CT-based AI metrics of the non-infected region and oxygenation, and their clinical outcomes were effectively improved after receiving invasive ventilation treatment. CONCLUSIONS A two-class LCA model based on AI quantification results of chest CT in the SP population particularly revealed clinical heterogeneity of lung function. Identifying the degree of match between ventilation and gas-exchange may help guide decisions about assisted ventilation.
Humans ; Tomography, X-Ray Computed/methods* ; Male ; Female ; Retrospective Studies ; Middle Aged ; Artificial Intelligence ; Aged ; Pneumonia/diagnosis* ; Latent Class Analysis ; Adult

Artificial intelligence (AI), particularly deep learning, has demonstrated remarkable performance in medical imaging across a variety of modalities, including X-ray, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, positron emission tomography (PET), and pathological imaging. However, most existing state-of-the-art AI techniques are task-specific and focus on a limited range of imaging modalities. Compared to these task-specific models, emerging foundation models represent a significant milestone in AI development. These models can learn generalized representations of medical images and apply them to downstream tasks through zero-shot or few-shot fine-tuning. Foundation models have the potential to address the comprehensive and multifactorial challenges encountered in clinical practice. This article reviews the clinical applications of both task-specific and foundation models, highlighting their differences, complementarities, and clinical relevance. We also examine their future research directions and potential challenges. Unlike the replacement relationship seen between deep learning and traditional machine learning, task-specific and foundation models are complementary, despite inherent differences. While foundation models primarily focus on segmentation and classification, task-specific models are integrated into nearly all medical image analyses. However, with further advancements, foundation models could be applied to other clinical scenarios. In conclusion, all indications suggest that task-specific and foundation models, especially the latter, have the potential to drive breakthroughs in medical imaging, from image processing to clinical workflows.
Humans ; Artificial Intelligence ; Deep Learning ; Diagnostic Imaging/methods* ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed ; Positron-Emission Tomography

OBJECTIVES: We present a new low-dose CT reconstruction method using sub-pixel and anisotropic diffusion. METHODS: The sub-pixel intensity values and their second-order differences were obtained using linear interpolation techniques, and the new gradient information was then embedded into an anisotropic diffusion process, which was introduced into a penalty-weighted least squares model to reduce the noise in low-dose CT projection data. The high-quality CT image was finally reconstructed using the classical filtered back-projection (FBP) algorithm from the estimated data. RESULTS: In the Shepp-Logan phantom experiments, the structural similarity (SSIM) index of the CT image reconstructed by the proposed algorithm, as compared with FBP, PWLS-Gibbs and PWLS-TV algorithms, was increased by 28.13%, 5.49%, and 0.91%, the feature similarity (FSIM) index was increased by 21.08%, 1.78%, and 1.36%, and the root mean square error (RMSE) was reduced by 69.59%, 18.96%, and 3.90%, respectively. In the digital XCAT phantom experiments, the SSIM index of the CT image reconstructed by the proposed algorithm, as compared with FBP, PWLS-Gibbs and PWLS-TV algorithms, was increased by 14.24%, 1.43% and 7.89%, the FSIM index was increased by 9.61%, 1.78% and 5.66%, and the RMSE was reduced by 26.88%, 9.41% and 18.39%, respectively. In clinical experiments, the SSIM index of the image reconstructed using the proposed algorithm was increased by 19.24%, 15.63% and 3.68%, the FSIM index was increased by 4.30%, 2.92% and 0.43%, and the RMSE was reduced by 44.60%, 36.84% and 15.22% in comparison with FBP, PWLS-Gibbs and PWLS-TV algorithms, respectively. CONCLUSIONS The proposed method can effectively reduce the noises and artifacts while maintaining the structural details in low-dose CT images.
Tomography, X-Ray Computed/methods* ; Algorithms ; Phantoms, Imaging ; Anisotropy ; Image Processing, Computer-Assisted/methods* ; Humans ; Radiation Dosage

OBJECTIVES: We propose a novel deep learning segmentation algorithm (DSRF) based on multi-scale supervision and residual feedback strategy for precise segmentation of the optic chiasm and optic nerves in CT images of nasopharyngeal carcinoma (NPC) patients. METHODS: We collected 212 NPC CT images and their ground truth labels from SegRap2023, StructSeg2019 and HaN-Seg2023 datasets. Based on a hybrid pooling strategy, we designed a decoder (HPS) to reduce small organ feature loss during pooling in convolutional neural networks. This decoder uses adaptive and average pooling to refine high-level semantic features, which are integrated with primary semantic features to enable network learning of finer feature details. We employed multi-scale deep supervision layers to learn rich multi-scale and multi-level semantic features under deep supervision, thereby enhancing boundary identification of the optic chiasm and optic nerves. A residual feedback module that enables multiple iterations of the network was designed for contrast enhancement of the optic chiasm and optic nerves in CT images by utilizing information from fuzzy boundaries and easily confused regions to iteratively refine segmentation results under supervision. The entire segmentation framework was optimized with the loss from each iteration to enhance segmentation accuracy and boundary clarity. Ablation experiments and comparative experiments were conducted to evaluate the effectiveness of each component and the performance of the proposed model. RESULTS: The DSRF algorithm could effectively enhance feature representation of small organs to achieve accurate segmentation of the optic chiasm and optic nerves with an average DSC of 0.837 and an ASSD of 0.351. Ablation experiments further verified the contributions of each component in the DSRF method. CONCLUSIONS The proposed deep learning segmentation algorithm can effectively enhance feature representation to achieve accurate segmentation of the optic chiasm and optic nerves in CT images of NPC.
Humans ; Tomography, X-Ray Computed/methods* ; Optic Chiasm/diagnostic imaging* ; Optic Nerve/diagnostic imaging* ; Algorithms ; Nasopharyngeal Carcinoma ; Deep Learning ; Nasopharyngeal Neoplasms/diagnostic imaging* ; Neural Networks, Computer ; Image Processing, Computer-Assisted/methods*

OBJECTIVES: We propose a low-dose CT image restoration method based on central guidance and alternating optimization (FedGP). METHODS: The FedGP framework revolutionizes the traditional federated learning model by adopting a structure without a fixed central server, where each institution alternatively serves as the central server. This method uses an institution-modulated CT image restoration network as the core of client-side local training. Through a federated learning approach of central guidance and alternating optimization, the central server leverages local labeled data to guide client-side network training to enhance the generalization capability of the CT imaging model across multiple institutions. RESULTS: In the low-dose and sparse-view CT image restoration tasks, the FedGP method showed significant advantages in both visual and quantitative evaluation and achieved the highest PSNR (40.25 and 38.84), the highest SSIM (0.95 and 0.92), and the lowest RMSE (2.39 and 2.56). Ablation study of FedGP demonstrated that compared with FedGP(w/o GP) without central guidance, the FedGP method better adapted to data heterogeneity across institutions, thus ensuring robustness and generalization capability of the model in different imaging conditions. CONCLUSIONS FedGP provides a more flexible FL framework to solve the problem of CT imaging heterogeneity and well adapts to multi-institutional data characteristics to improve generalization ability of the model under diverse imaging geometric configurations.
Tomography, X-Ray Computed/methods* ; Humans ; Radiation Dosage ; Image Processing, Computer-Assisted/methods* ; Algorithms

OBJECTIVES: To establish a three-dimensional (3D) visualization-based classification of the left hepatic portal vein (LHPV) and left hepatic vein (LHV) systems using 3D reconstruction technology to facilitate precise segmental/subsegmental resection of left liver lesions. METHODS: Thin-slice contrast-enhanced CT datasets from 244 patients were reconstructed using MI-3DV Works software. The spatial anatomy (origins, branching patterns, and spatial relationships) of the LHPV and LHV branches was analyzed to determine their 3D classifications and segmental liver divisions for guiding surgical planning for anatomical left liver resections. RESULTS: The 3D models of the third- and fourth-order branches of the LHPV and LHV were successfully reconstructed for all the 244 patients. Two types of the LHPV system were identified, where the LHPV either had independent origins [242 cases (99.1%)] or had right anterior portal branches arising from the LHPV trunk [2 cases (0.9%)]. 3D classifications identified two types of the Segment II of the LHPV (based on branch number), 3 types of the Segment III (by spatial distribution of the branches), compact vs dispersed types of the left lateral lobe (determined by Segment II/III branches proximity), 3 types of the Segment IV (by branch number and origin), and 3 types the fourth hilar vessels (transverse branches of the left portal vein) for their supplied segments. The LHV system had two drainage types into the inferior vena cava, and the umbilical fissure veins were classified into 3 types by drainage patterns and distance to the venous roots. These classifications combined with liver segmentations allowed individualized surgical planning for segment-specific resections. CONCLUSIONS The 3D classification of the LHPV and LHV provides valuable clinical guidance for precise anatomical resections of left liver lesions using liver segments or subsegments as anatomical units to enhance surgical accuracy and improve the outcomes of hepatobiliary surgery.
Humans ; Hepatectomy/methods* ; Imaging, Three-Dimensional ; Hepatic Veins/anatomy & histology* ; Portal Vein/anatomy & histology* ; Liver/surgery* ; Liver Neoplasms/blood supply* ; Tomography, X-Ray Computed ; Female

OBJECTIVES: To construct a bone tumor classification model based on feature decoupling and fusion for processing modality loss and fusing multimodal information to improve classification accuracy. METHODS: A decoupling completion module was designed to extract local and global bone tumor image features from available modalities. These features were then decomposed into shared and modality-specific features, which were used to complete the missing modality features, thereby reducing completion bias caused by modality differences. To address the challenge of modality differences that hinder multimodal information fusion, a cross-attention-based fusion module was introduced to enhance the model's ability to learn cross-modal information and fully integrate specific features, thereby improving the accuracy of bone tumor classification. RESULTS: The experiment was conducted using a bone tumor dataset collected from the Third Affiliated Hospital of Southern Medical University for training and testing. Among the 7 available modality combinations, the proposed method achieved an average AUC, accuracy, and specificity of 0.766, 0.621, and 0.793, respectively, which represent improvements of 2.6%, 3.5%, and 1.7% over existing methods for handling missing modalities. The best performance was observed when all the modalities were available, resulting in an AUC of 0.837, which still reached 0.826 even with MRI alone. CONCLUSIONS The proposed method can effectively handle missing modalities and successfully integrate multimodal information, and show robust performance in bone tumor classification under various complex missing modality scenarios.
Humans ; Bone Neoplasms/diagnosis* ; Multimodal Imaging/methods* ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed ; Image Processing, Computer-Assisted/methods* ; Algorithms

Tongue squamous cell carcinoma (TSCC) is a prevalent malignancy that afflicts the head and neck area and presents a high incidence of metastasis and invasion. Accurate diagnosis and effective treatment are essential for enhancing the quality of life and the survival rates of TSCC patients. The current treatment modalities for TSCC frequently suffer from a lack of specificity and efficacy. Nanoparticles with diagnostic and photothermal therapeutic properties may offer a new approach for the targeted therapy of TSCC. However, inadequate accumulation of photosensitizers at the tumor site diminishes the efficacy of photothermal therapy (PTT). This study modified gold nanodots (AuNDs) with the TSCC-targeting peptide HN-1 to improve the selectivity and therapeutic effects of PTT. The Au-HN-1 nanosystem effectively targeted the TSCC cells and was rapidly delivered to the tumor tissues compared to the AuNDs. The enhanced accumulation of photosensitizing agents at tumor sites achieved significant PTT effects in a mouse model of TSCC. Moreover, owing to its stable long-term fluorescence and high X-ray attenuation coefficient, the Au-HN-1 nanosystem can be used for fluorescence and computed tomography imaging of TSCC, rendering it useful for early tumor detection and accurate delineation of surgical margins. In conclusion, Au-HN-1 represents a promising nanomedicine for imaging-based diagnosis and targeted PTT of TSCC.
Tongue Neoplasms/diagnostic imaging* ; Carcinoma, Squamous Cell/diagnostic imaging* ; Animals ; Gold/chemistry* ; Mice ; Photothermal Therapy/methods* ; Tomography, X-Ray Computed ; Photosensitizing Agents ; Metal Nanoparticles ; Humans ; Cell Line, Tumor

OBJECTIVES: This retrospective study aimed to investigate factors influencing positional changes of the condyle and temporomandibular joint (TMJ) following mandibular defect reconstruction with bone flaps, and to evaluate the biomechanical impacts of flap reconstruction on condylar positioning, thereby providing evidence for optimizing surgical protocols and TMJ functional rehabilitation. METHODS: A retrospective study was conducted on 90 patients undergoing mandibular segmental resection with immediate bone flap reconstruction at Guizhou Medical University Affiliated Stomatological Hospital (June 2019 to May 2024). After strict screening, 50 cases with complete data were analyzed. Clinical parameters (defect size, location, reconstruction method) and craniofacial CT scans at four timepoints [preoperative (T0), 7-10 days (T1), 3 months (T2), and 6 months (T3) postoperatively] were collected. Mimics 20 software facilitated 3D reconstruction for measuring TMJ anterior/posterior/superior joint spaces (Kamelchuk method) and calculating condylar position via the Pullinger index [Ln (posterior/anterior space)]. Vitral and Krisjane methods quantified mandibular linear parameters (ramus length, condylar pole distances to the sagittal plane, angulation) and glenoid fossa morphology. Statistical analyses were performed using SPSS 21.0. RESULTS: Mandibular defect size and location were significant factors influencing postoperative condylar position changes (P<0.05). Compared to preoperative measurements, postoperative condylar anterior, posterior, and superior joint spaces were significantly increased (P<0.001). The most pronounced anterior condylar displacement occurred within 7-10 days postoperatively (P<0.05). In patients with condyle resection, postoperative joint space and angle changes were significant; in patients with condyle preservation, only superior and anterior joint space changes were statistically significant (P<0.05). Additionally, from T1 to T2, the changes in condylar medial-lateral distance, superior joint space, and anterior joint space were negatively correlated with the preoperative condylar position. Compared with preoperative,in the T0-T1 period, condylar medial-lateral distance, posterior joint space, and articular tubercle angle changes were significantly negatively correlated with time (P<0.05). Notably, the angle between the condylar long axis and the coronal axis showed a sustained negative trend from T1 to T3 (P<0.05). CONCLUSIONS Condylar position changes after mandibular defect repair with bone flap reconstruction are associated with the size and location of the defect. Additionally, adaptive remodeling of the temporomandibular joint (TMJ) joint space occurs postoperatively. The phenomenon of anterior displacement of the condyle in the early postoperative period (7-10 days) shows a trend of reduction with prolonged follow-up time, and further sample size research is needed.
Humans ; Retrospective Studies ; Temporomandibular Joint/surgery* ; Mandibular Condyle/surgery* ; Male ; Female ; Adult ; Middle Aged ; Mandibular Reconstruction/methods* ; Mandible/surgery* ; Surgical Flaps ; Tomography, X-Ray Computed ; Young Adult ; Biomechanical Phenomena ; Aged ; Adolescent ; Imaging, Three-Dimensional

A nodule in the right middle lobe of the lung was treated by a combination of cone-beam CT,three-dimensional registration for fusion imaging,and electromagnetic navigation bronchoscopy-guided thermal ablation.The procedure lasted for 90 min,with no significant bleeding observed under the bronchoscope.The total radiation dose during the operation was 384 mGy.The patient recovered well postoperatively,with only a small amount of blood in the sputum and no pneumothorax or other complications.A follow-up chest CT on the first day post operation showed that the ablation area completely covered the lesion,and the patient was discharged successfully.
Humans ; Bronchoscopy/methods* ; Catheter Ablation/methods* ; Cone-Beam Computed Tomography ; Electromagnetic Phenomena ; Imaging, Three-Dimensional ; Lung Neoplasms/diagnostic imaging* ; Tomography, X-Ray Computed