BACKGROUND AND OBJECTIVES

The Philippines faces challenges in the screening of tuberculosis (TB), one of them being the shortage in the health workforce who are skilled and allowed to screen TB. Deep learning neural networks (DLNNs) have shown potential in the TB screening process utilizing chest radiographs (CXRs). However, local studies on AIbased TB screening are limited. This study evaluated qXR3.0 technology's diagnostic performance for TB screening in Filipino adults aged 15 and older. Specifically, we evaluated the specificity and sensitivity of qXR3.0 compared to radiologists' impressions and determined whether it meets the World Health Organization (WHO) standards.

A prospective cohort design was used to perform a study on comparing screening and diagnostic accuracies of qXR3.0 and two radiologist gradings in accordance with the Standards for Reporting Diagnostic Accuracy (STARD). Subjects from two clinics in Metro Manila which had qXR 3.0 seeking consultation at the time of study were invited to participate to have CXRs and sputum collected. Radiologists' and qXR3.0 readings and impressions were compared with respect to the reference standard Xpert MTB/RiF assay. Diagnostic accuracy measures were calculated.

With 82 participants, qXR3.0 demonstrated 100% sensitivity and 72.7% specificity with respect to the reference standard. There was a strong agreement between qXR3.0 and radiologists' readings as exhibited by the 0.7895 (between qXR 3.0 and CXRs read by at least one radiologist), 0.9362 (qXR 3.0 and CXRs read by both radiologists), and 0.9403 (qXR 3.0 and CXRs read as not suggestive of TB by at least one radiologist) concordance indices.

qXR3.0 demonstrated high sensitivity to identify presence of TB among patients, and meets the WHO standard of at least 70% specificity for detecting true TB infection. This shows an immense potential for the tool to supplement the shortage of radiologists for TB screening in the country. Future research directions may consider larger sample sizes to confirm these findings and explore the economic value of mainstream adoption of qXR 3.0 for TB screening.

Testicular histology based on testicular biopsy is an important factor for determining appropriate testicular sperm extraction surgery and predicting sperm retrieval outcomes in patients with azoospermia. Therefore, we developed a deep learning (DL) model to establish the associations between testicular grayscale ultrasound images and testicular histology. We retrospectively included two-dimensional testicular grayscale ultrasound from patients with azoospermia (353 men with 4357 images between July 2017 and December 2021 in The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China) to develop a DL model. We obtained testicular histology during conventional testicular sperm extraction. Our DL model was trained based on ultrasound images or fusion data (ultrasound images fused with the corresponding testicular volume) to distinguish spermatozoa presence in pathology (SPP) and spermatozoa absence in pathology (SAP) and to classify maturation arrest (MA) and Sertoli cell-only syndrome (SCOS) in patients with SAP. Areas under the receiver operating characteristic curve (AUCs), accuracy, sensitivity, and specificity were used to analyze model performance. DL based on images achieved an AUC of 0.922 (95% confidence interval [CI]: 0.908-0.935), a sensitivity of 80.9%, a specificity of 84.6%, and an accuracy of 83.5% in predicting SPP (including normal spermatogenesis and hypospermatogenesis) and SAP (including MA and SCOS). In the identification of SCOS and MA, DL on fusion data yielded better diagnostic performance with an AUC of 0.979 (95% CI: 0.969-0.989), a sensitivity of 89.7%, a specificity of 97.1%, and an accuracy of 92.1%. Our study provides a noninvasive method to predict testicular histology for patients with azoospermia, which would avoid unnecessary testicular biopsy.
Humans ; Male ; Azoospermia/diagnostic imaging* ; Deep Learning ; Testis/pathology* ; Retrospective Studies ; Adult ; Ultrasonography/methods* ; Sperm Retrieval ; Sertoli Cell-Only Syndrome/diagnostic imaging*

With the development of artificial intelligence technology and the growing demand for minimally invasive surgery, the intelligentization of minimally invasive surgery has become a current research hotspot. Surgical instrument segmentation is a highly promising technology that can enhance the performance of minimally invasive endoscopic imaging systems, surgical video analysis systems, and other related systems. This article summarizes the semantic and instance segmentation methods of minimally invasive surgical instruments based on deep learning, deeply analyzes the supervision methods of training algorithms, network structure improvements, and attention mechanisms, and then discusses the methods based on the Segment Anything Model. Given that deep learning methods have extremely high requirements for data, current data augmentation methods have also been explored. Finally, a summary and outlook on instrument segmentation technology are provided.
Artificial Intelligence ; Minimally Invasive Surgical Procedures/instrumentation* ; Algorithms ; Deep Learning ; Humans ; Image Processing, Computer-Assisted

The clinical application of binocular endoscope relies primarily on the visual system of physicians to create a three-dimensional effect, but it cannot provide accurate depth information. The utilization of 3D reconstruction technology in binocular endoscopy can facilitate the recovery of image depth information, and the application of deep learning-based 3D reconstruction technology can significantly improve the accuracy and real-time performance of reconstruction results, making it widely applicable in the realm of minimally invasive surgery. This paper aims to explore the key technologies and implementation methods of deep learning based 3D reconstruction for binocular endoscopic images, and seeks to outline strategies for enhancing the quality of 3D reconstruction in endoscopic images, providing guidance for sustainable development of binocular endoscopic image reconstruction technology in clinical settings. This will assist in the application of minimally invasive surgery and contribute to meeting the demands of precision medicine.
Deep Learning ; Imaging, Three-Dimensional/methods* ; Humans ; Endoscopy/methods* ; Image Processing, Computer-Assisted/methods* ; Minimally Invasive Surgical Procedures

Computer-assisted methods for pathological image analysis can improve doctor's efficiency of image reading and diagnostic accuracy, effectively addressing the shortage of pathology diagnostic manpower. With the rapid development of artificial intelligence and digital pathology, deep learning technology has spurred a wealth of research in the field of histopathology. This article reviews the various applications of deep learning in digital pathological image analysis, such as pathological image segmentation, cancer auxiliary diagnosis, and cancer prognosis prediction, and discusses the challenges and solutions in its application. Furthermore, it predicts future trends in deep learning for pathological image analysis and proposes potential research directions.
Deep Learning ; Humans ; Image Processing, Computer-Assisted/methods* ; Artificial Intelligence ; Neoplasms

OBJECTIVES: Renal cell carcinoma (RCC) is a malignant renal tumor that poses a significant threat to patient health. Accurate preoperative pathological grading plays a crucial role in determining the appropriate treatment for this disease. Currently, deep learning technology has become an important method for pathological grading of RCC. However, existing methods primarily rely on single-phase computed tomography (CT) imaging for analysis and prediction, which has limitations such as missing small lesions, one-sided evaluation, and local focusing issues. Therefore, this study proposes a multi-modal deep learning algorithm that integrates multi-phase enhanced CT images with clinical variable data, aiming to provide a basis for predicting the pathological grading of RCC. METHODS: First, the algorithm took four-phase enhanced CT images from the plain scan, arterial phase, venous phase, and delayed phase, along with clinical variables, as inputs. Then, an embedding encoding module was used to extract heterogeneous information from the clinical variables, and a 3-dimensional (3D) ResNet50 model was employed to capture spatial information from the multi-phase enhanced CT image data. Finally, a Fusion module deeply integrated the feature information from clinical variables and each phase's CT image features, further utilizing a cross-self-attention mechanism to achieve multi-phase feature fusion. This approach comprehensively captures the deep semantic information from the patient data, fully leveraging the complementary advantages of multi-modal and multi-phase data. To validate the effectiveness of the proposed method, a total of 1 229 RCC patients were approved by ethics review were included to train the model. RESULTS: Experimental results demonstrated superior performance compared to traditional radiomics and state-of-the-art deep learning methods, achieving an accuracy of 83.87%, a recall rate of 95.04%, and an F1-score of 82.23%. CONCLUSIONS The proposed algorithm exhibits strong stability and sensitivity, significantly enhancing the predictive performance of RCC pathological grading. It offers a novel approach for accurate RCC diagnosis and personalized treatment planning.
Humans ; Carcinoma, Renal Cell/pathology* ; Deep Learning ; Kidney Neoplasms/diagnostic imaging* ; Tomography, X-Ray Computed/methods* ; Algorithms ; Neoplasm Grading ; Male ; Female ; Middle Aged

Antibodies currently comprise the predominant treatment modality for a variety of diseases; therefore, optimizing their properties rapidly and efficiently is an indispensable step in antibody-based drug development. Inspired by the great success of artificial intelligence-based algorithms, especially deep learning-based methods in the field of biology, various computational methods have been introduced into antibody optimization to reduce costs and increase the success rate of lead candidate generation and optimization. Herein, we briefly review recent progress in deep learning-based antibody optimization, focusing on the available datasets and algorithm input data types that are crucial for constructing appropriate deep learning models. Furthermore, we discuss the current challenges and potential solutions for the future development of general-purpose deep learning algorithms in antibody optimization.
Deep Learning ; Humans ; Antibodies/chemistry* ; Algorithms ; Artificial Intelligence ; Drug Development

The glucagon receptor (GCGR) is a critical target for the treatment of metabolic disorders such as Type 2 Diabetes Mellitus (T2DM) and obesity. Activation of GCGR enhances systemic insulin sensitivity through paracrine stimulation of insulin secretion, presenting a promising avenue for treatment. However, the discovery of effective GCGR agonists remains a challenging and resource-intensive process, often requiring time-consuming wet-lab experiments to synthesize and screen potential compounds. Recent advances in artificial intelligence technologies have demonstrated great potential in accelerating drug discovery by streamlining screening and efficiently predicting bioactivity. In the present work, we propose DeepGCGR, a two-layer deep learning model that leverages graph convolutional networks (GCN) integrated with a multiple attention mechanism to expedite the identification of GCGR agonists. In the first layer, the model predicts the bioactivity of various compounds against GCGR, efficiently filtering large chemical libraries to identify promising candidates. In the second layer, DeepGCGR classifies high bioactive compounds based on their functional effects on GCGR signaling, identifying those with potential agonistic or antagonistic effects. Moreover, DeepGCGR was specifically applied to identify novel GCGR-regulating compounds for the treatment of T2DM from natural products derived from traditional Chinese medicine (TCM). The proposed method will not only offer an effective strategy for discovering GCGR-targeting compounds with functional activation properties but also provide new insights into the development of T2DM therapeutics.
Deep Learning ; Drug Discovery/methods* ; Humans ; Diabetes Mellitus, Type 2/metabolism* ; Medicine, Chinese Traditional ; Drugs, Chinese Herbal/pharmacology*

Artificial intelligence (AI) has emerged as a transformative technology in accelerating drug discovery and development within natural medicines research. Natural medicines, characterized by their complex chemical compositions and multifaceted pharmacological mechanisms, demonstrate widespread application in treating diverse diseases. However, research and development face significant challenges, including component complexity, extraction difficulties, and efficacy validation. AI technology, particularly through deep learning (DL) and machine learning (ML) approaches, enables efficient analysis of extensive datasets, facilitating drug screening, component analysis, and pharmacological mechanism elucidation. The implementation of AI technology demonstrates considerable potential in virtual screening, compound optimization, and synthetic pathway design, thereby enhancing natural medicines' bioavailability and safety profiles. Nevertheless, current applications encounter limitations regarding data quality, model interpretability, and ethical considerations. As AI technologies continue to evolve, natural medicines research and development will achieve greater efficiency and precision, advancing both personalized medicine and contemporary drug development approaches.
Biological Products/pharmacology* ; Artificial Intelligence ; Humans ; Drug Discovery/methods* ; Machine Learning ; Deep Learning

Objective:To development a deep learning（DL） model based on conventional MRI for automatic segmentation and differential diagnosis of nasopharyngeal carcinoma（NPC） and nasopharyngeal lymphoma（NPL）. Methods:The retrospective study included 142 patients with NPL and 292 patients with NPC who underwent conventional MRI at Renmin Hospital of Wuhan University from June 2012 to February 2023. MRI from 80 patients were manually segmented to train the segmentation model. The automatically segmented regions of interest（ROIs） formed four datasets: T1 weighted images（T1WI）, T2 weighted images（T2WI）, T1 weighted contrast-enhanced images（T1CE）, and a combination of T1WI and T2WI. The ImageNet-pretrained ResNet101 model was fine-tuned for the classification task. Statistical analysis was conducted using SPSS 22.0. The Dice coefficient loss was used to evaluate performance of segmentation task. Diagnostic performance was assessed using receiver operating characteristic（ROC） curves. Gradient-weighted class activation mapping（Grad-CAM） was imported to visualize the model's function. Results:The DICE score of the segmentation model reached 0.876 in the testing set. The AUC values of classification models in testing set were as follows: T1WI: 0.78（95%CI 0.67-0.81）, T2WI: 0.75（95%CI 0.72-0.86）, T1CE: 0.84（95%CI 0.76-0.87）, and T1WI+T2WI: 0.93（95%CI 0.85-0.94）. The AUC values for the two clinicians were 0.77（95%CI 0.72-0.82） for the junior, and 0.84（95%CI 0.80-0.89） for the senior. Grad-CAM analysis revealed that the central region of the tumor was highly correlated with the model's classification decisions, while the correlation was lower in the peripheral regions. Conclusion:The deep learning model performed well in differentiating NPC from NPL based on conventional MRI. The T1WI+T2WI combination model exhibited the best performance. The model can assist in the early diagnosis of NPC and NPL, facilitating timely and standardized treatment, which may improve patient prognosis.
Humans ; Nasopharyngeal Carcinoma/diagnostic imaging* ; Deep Learning ; Magnetic Resonance Imaging ; Retrospective Studies ; Nasopharyngeal Neoplasms/diagnostic imaging* ; Lymphoma/diagnostic imaging* ; Diagnosis, Differential ; ROC Curve ; Male ; Female ; Middle Aged ; Adult