Detection Of Lung Tumors Research Articles

Plasmonic trimers designed as SERS-active chemical traps for subtyping of lung tumors

Plasmonic materials can generate strong electromagnetic fields to boost the Raman scattering of surrounding molecules, known as surface-enhanced Raman scattering. However, these electromagnetic fields are heterogeneous, with only molecules located at the ‘hotspots’, which account for ≈ 1% of the surface area, experiencing efficient enhancement. Herein, we propose patterned plasmonic trimers, consisting of a pair of plasmonic dimers at the bilateral sides and a trap particle positioned in between, to address this challenge. The trimer configuration selectively directs probe molecules to the central traps where ‘hotspots’ are located through chemical affinity, ensuring a precise spatial overlap between the probes and the location of maximum field enhancement. We investigate the Raman enhancement of the Au@Al2O3-Au-Au@Al2O3 trimers, achieving a detection limit of 10−14 M of 4-methylbenzenethiol, 4-mercaptopyridine, and 4-aminothiophenol. Moreover, single-molecule SERS sensitivity is demonstrated by a bi-analyte method. Benefiting from this sensitivity, our approach is employed for the early detection of lung tumors using fresh tissues. Our findings suggest that this approach is sensitive to adenocarcinoma but not to squamous carcinoma or benign cases, offering insights into the differentiation between lung tumor subtypes.

Study of a Deep Convolution Network with Enhanced Region Proposal Network in the Detection of Cancerous Lung Tumors.

A deep convolution network that expands on the architecture of the faster R-CNN network is proposed. The expansion includes adapting unsupervised classification with multiple backbone networks to improve the Region Proposal Network in order to improve accuracy and sensitivity in detecting minute changes in images. The efficiency of the proposed architecture is investigated by applying it to the detection of cancerous lung tumors in CT (computed tomography) images. This investigation used a total of 888 images from the LUNA16 dataset, which contains CT images of both cancerous and non-cancerous tumors of various sizes. These images are divided into 80% and 20%, which are used for training and testing, respectively. The result of the investigation through the experiment is that the proposed deep-learning architecture could achieve an accuracy rate of 95.32%, a precision rate of 94.63%, a specificity of 94.84%, and a high sensitivity of 96.23% using the LUNA16 images. The result shows an improvement compared to a reported accuracy of 93.6% from a previous study using the same dataset.

An enhanced Garter Snake Optimization-assisted deep learning model for lung cancer segmentation and classification using CT images

An early detection of lung tumors is critical for better treatment results, and CT scans can reveal lumps in the lungs which are too small to be picked up by conventional X-rays. CT imaging has advantages, but it also exposes a person to radiation from ions, which raises the possibility of malignancy, particularly when the imaging procedure is done. Access to expensive-quality CT scans and the related sophisticated analytic tools might be restricted in environments with fewer resources due to their high cost and limited availability. It will need an array of creative technological innovations to overcome such weaknesses. This paper aims to design a heuristic and deep learning-aided lung cancer classification using CT images. The collected images are undergone for segmentation, which is performed by Shuffling Atrous Convolutional (SAC) based ResUnet++ (SACRUnet++). Finally, the lung cancer classification is performed by the Adaptive Residual Attention Network (ARAN) by inputting the segmented images. Here the parameters of ARAN are optimally tuned using the Improved Garter Snake Optimization Algorithm (IGSOA). The developed lung cancer classification performance is compared to conventional lung cancer classification models and it showed high accuracy.

Novel locally nebulized indocyanine green for simultaneous identification of tumor margin and intersegmental plane.

Segmentectomy, recommended for early-stage lung cancer or compromised lung function, demands precise tumor detection and intersegmental plane identification. While indocyanine green (ICG) commonly aids in these aspects using near-infrared imaging, its separate administrations through different routes and times can lead to complications and patient anxiety. This study aims to develop a lung-specific delivery method by nebulizing low-dose ICG to targeted lung segments, allowing simultaneous detection of lung tumors and intersegmental planes across diverse animal models. To optimizing the dose of ICG for lung tumor and interlobar fissure detection, different doses of ICG (0.25, 0.1, and 0.05mg/kg) were nebulized to rabbit lung tumor models. The distribution of locally nebulized ICG in targeted segments was studied to evaluate the feasibility of detecting lung tumor and intersegmental planes in canine lung pseudotumor models. Near-infrared fluorescence imaging demonstrated clear visualization of lung tumor margin and interlobar fissure using local nebulization of 0.1mg/kg ICG for only 4min during surgery in the rabbit models. In the canine model, the local nebulization of 0.05mg/kg of ICG into the target segment enabled clear visualization of pseudotumor and intersegmental planes for 30min. This innovative approach achieves a reduction in ICG dose and prolonged the visualization time of the intersegmental plane and effectively eliminates the need for the hurried marking of tumors and intersegmental planes. The authors anticipate that lung-specific delivery of ICG will prove valuable for image-guided limited resection of lung tumors in clinical practice.

Development and Validation of a Deep Learning Model for Histopathological Slide Analysis in Lung Cancer Diagnosis.

Lung cancer is the leading cause of cancer-related deaths worldwide. Two of the crucial factors contributing to these fatalities are delayed diagnosis and suboptimal prognosis. The rapid advancement of deep learning (DL) approaches provides a significant opportunity for medical imaging techniques to play a pivotal role in the early detection of lung tumors and subsequent monitoring during treatment. This study presents a DL-based model for efficient lung cancer detection using whole-slide images. Our methodology combines convolutional neural networks (CNNs) and separable CNNs with residual blocks, thereby improving classification performance. Our model improves accuracy (96% to 98%) and robustness in distinguishing between cancerous and non-cancerous lung cell images in less than 10 s. Moreover, the model's overall performance surpassed that of active pathologists, with an accuracy of 100% vs. 79%. There was a significant linear correlation between pathologists' accuracy and years of experience (r Pearson = 0.71, 95% CI 0.14 to 0.93, p = 0.022). We conclude that this model enhances the accuracy of cancer detection and can be used to train junior pathologists.

DLCT LUNG Detect Net: Leveraging Deep Learning for Lung Tumor Detection in CT scans

Lung cancer is a critical global health concern, necessitating precise early diagnosis and intervention for better patient outcomes. Computed Tomography (CT) scans are pivotal in lung cancer detection, and leveraging advanced technology is crucial. This study introduces "DLCTLungDetectNet," a Convolutional Neural Network (CNN) based deep learning framework, with a focus on early lung cancer detection using CT scan images. The core innovation lies in the integration of the robust "FusionNet," a hybrid model amalgamating feature from ResNet50 and InceptionV3. We conduct a comprehensive comparative analysis, showcasing the superior performance of DLCTLungDetectNet over established architectures such as VGG16, VGG19, and Inception v3. Rigorous evaluation based on standard metrics substantiates DLCTLungDetectNet's high accuracy, precision, Area Under Curve (AUC), and F1 score. This research not only highlights the potential of deep learning in enhancing lung cancer diagnosis but also establishes a benchmark, showcasing the efficacy of the FusionNet hybrid model for achieving superior accuracy in automated lung tumor detection.

Non-invasive detection of orthotopic human lung tumors by microRNA expression profiling of mouse exhaled breath condensates and exhaled extracellular vesicles.

The lung is the second most frequent site of metastatic dissemination. Early detection is key to improving survival. Given that the lung interfaces with the external environment, the collection of exhaled breath condensate (EBC) provides the opportunity to obtain biological material including exhaled miRNAs that originate from the lung. In this proof-of-principal study, we used the highly metastatic MDA-MB-231 subline 3475 breast cancer cell line (LM-3475) to establish an orthotopic lung tumor-bearing mouse model and investigate non-invasive detection of lung tumors by analysis of exhaled miRNAs. We initially conducted miRNA NGS and qPCR validation analyses on condensates collected from unrestrained animals and identified significant miRNA expression differences between the condensates of lung tumor-bearing and control mice. To focus our purification of EBC and evaluate the origin of these differentially expressed miRNAs, we developed a system to collect EBC directly from the nose and mouth of our mice. Using nanoparticle distribution analyses, TEM, and ONi super-resolution nanoimaging, we determined that human tumor EVs could be increasingly detected in mouse EBC during the progression of secondary lung tumors. Using our customizable EV-CATCHER assay, we purified human tumor EVs from mouse EBC and demonstrated that the bulk of differentially expressed exhaled miRNAs originate from lung tumors, which could be detected by qPCR within 1 to 2 weeks after tail vein injection of the metastatic cells. This study is the first of its kind and demonstrates that lung tumor EVs are exhaled in mice and provide non-invasive biomarkers for detection of lung tumors.

Experimental Detection of Early-Stage Lung and Skin Tumors Based on Super Wideband Imaging

Experimental Detection of Early-Stage Lung and Skin Tumors Based on Super Wideband Imaging

Innovative Fusion Approach for Lung Tumor Detection: Double Deep CNN Empowered by SVM Integration in CT Scan Images

Innovative Fusion Approach for Lung Tumor Detection: Double Deep CNN Empowered by SVM Integration in CT Scan Images

KÜÇÜK HÜCRELİ DIŞI AKCİĞER TÜMÖRLERİNİN DEĞERLENDİRİLMESİNDE Tc-99m MIBI SPECT’İN YERİ

Objective: In this study, it was aimed to investigate the role of Tc-99m MIBI scintigraphy in the diagnosis and evaluation of non-small cell lung cancers and also to evaluate the use of MIBI scintigraphy and MDP bone scintigraphy together in the detection of metastases. Methods: For this purpose, 30 patients with non-small cell lung cancer and 15 patients with benign lesions other than lung cancer (as a control group) were included in the study. By applying Tc-99m MIBI and Tc99m MDP scintigraphy to whole patients with benign and malign lung lesions, anterior-posterior planar and SPECT MIBI images were obtained. Bone scintigraphy was performed as whole-body images. Scintigraphic findings were compared with the histopathological diagnosis of patients. Data were analyzed statistically. Results: According to the result of the findings, the sensitivity, specificity, positive, predictive value, negative predictive value, and accuracy rate of Tc-99m MIBI scintigraphy in the detection of non-small cell lung tumors were determined as 87%, 93.5%, 96%, 77.7%, and 88.8%, respectively. Scintigraphic findings were evaluated quantitatively and visually. While there was no statistically significant difference between the early uptake rates of non-small cell lung tumors and benign lesions in the quantitative evaluation using planar MIBI images, late Tc-99m MIBI uptake rates were significantly higher in lesions with non-small cell lung tumors than that of benign lesions. However, it was detected that both the early and late uptake ratios were significantly higher in malign lesions than in benign lesions in the quantitative evaluation of SPECT images. It was observed that MIBI washout in benign lesions was higher than in malignant lesions in visual washout analysis. Additionally, multiple bone metastases were observed in 14 patients on the visual evaluation of bone scan of the patients with non-small cell lung cancer. Tc-99m MDP uptake was not observed in the malignant lung lesion region or in the soft tissue except the lesion. Conclusion: In light of the findings, it was concluded that Tc-99m MIBI scintigraphy is a noninvasive method with high sensitivity and specificity that can be used to differentiate non-small cell lung tumors from benign lesions.

Behind the Indolent Facade: Uncovering the Molecular Features and Malignancy Potential in Lung Minimally Invasive Adenocarcinoma by Single-Cell Transcriptomics.

The increased use of low-dose computed tomography screening has led to more frequent detection of early stage lung tumors, including minimally invasive adenocarcinoma (MIA). To unravel the intricacies of tumor cells and the immune microenvironment in MIA, this study performs a comprehensive single-cell transcriptomic analysis and profiles the transcriptomes of 156,447 cells from fresh paired MIA and invasive adenocarcinoma (IA) tumor samples, peripheral blood mononuclear cells, and adjacent normal tissue samples from three patients with synchronous multiple primary lung adenocarcinoma. This study highlights a connection and heterogeneity between the tumor ecosystem of MIA and IA. MIA tumor cells exhibited high expression of aquaporin-1 and angiotensin II receptor type 2 and a basal-like molecular character. Furthermore, it identifies that cathepsin B+ tumor-associated macrophages may over-activate CD8+ T cells in MIA, leading to an enrichment of granzyme K+ senescent CD8+ T cells, indicating the possibility of malignant progression behind the indolent appearance of MIA. These findings are further validated in 34 MIA and 35 IA samples by multiplexed immunofluorescence. These findings provide valuable insights into the mechanisms that maintain the indolent nature and prompt tumor progression of MIA and can be used to develop more effective therapeutic targets and strategies for MIA patients.

Weakly supervised segmentation models as explainable radiological classifiers for lung tumour detection on CT images

PurposeInterpretability is essential for reliable convolutional neural network (CNN) image classifiers in radiological applications. We describe a weakly supervised segmentation model that learns to delineate the target object, trained with only image-level labels (“image contains object” or “image does not contain object”), presenting a different approach towards explainable object detectors for radiological imaging tasks.MethodsA weakly supervised Unet architecture (WSUnet) was trained to learn lung tumour segmentation from image-level labelled data. WSUnet generates voxel probability maps with a Unet and then constructs an image-level prediction by global max-pooling, thereby facilitating image-level training. WSUnet’s voxel-level predictions were compared to traditional model interpretation techniques (class activation mapping, integrated gradients and occlusion sensitivity) in CT data from three institutions (training/validation: n = 412; testing: n = 142). Methods were compared using voxel-level discrimination metrics and clinical value was assessed with a clinician preference survey on data from external institutions.ResultsDespite the absence of voxel-level labels in training, WSUnet’s voxel-level predictions localised tumours precisely in both validation (precision: 0.77, 95% CI: [0.76–0.80]; dice: 0.43, 95% CI: [0.39–0.46]), and external testing (precision: 0.78, 95% CI: [0.76–0.81]; dice: 0.33, 95% CI: [0.32–0.35]). WSUnet’s voxel-level discrimination outperformed the best comparator in validation (area under precision recall curve (AUPR): 0.55, 95% CI: [0.49–0.56] vs. 0.23, 95% CI: [0.21–0.25]) and testing (AUPR: 0.40, 95% CI: [0.38–0.41] vs. 0.36, 95% CI: [0.34–0.37]). Clinicians preferred WSUnet predictions in most instances (clinician preference rate: 0.72 95% CI: [0.68–0.77]).ConclusionWeakly supervised segmentation is a viable approach by which explainable object detection models may be developed for medical imaging.Critical relevance statementWSUnet learns to segment images at voxel level, training only with image-level labels. A Unet backbone first generates a voxel-level probability map and then extracts the maximum voxel prediction as the image-level prediction. Thus, training uses only image-level annotations, reducing human workload. WSUnet’s voxel-level predictions provide a causally verifiable explanation for its image-level prediction, improving interpretability.Key points• Explainability and interpretability are essential for reliable medical image classifiers.• This study applies weakly supervised segmentation to generate explainable image classifiers.• The weakly supervised Unet inherently explains its image-level predictions at voxel level.Graphical

Machine Learning Model of ResNet50-Ensemble Voting for Malignant-Benign Small Pulmonary Nodule Classification on Computed Tomography Images.

The early detection of benign and malignant lung tumors enabled patients to diagnose lesions and implement appropriate health measures earlier, dramatically improving lung cancer patients' quality of living. Machine learning methods performed admirably when recognizing small benign and malignant lung nodules. However, exploration and investigation are required to fully leverage the potential of machine learning in distinguishing between benign and malignant small lung nodules. The aim of this study was to develop and evaluate the ResNet50-Ensemble Voting model for detecting the benign and malignant nature of small pulmonary nodules (<20 mm) based on CT images. In this study, 834 CT imaging data from 396 patients with small pulmonary nodules were gathered and randomly assigned to the training and validation sets in an 8:2 ratio. ResNet50 and VGG16 algorithms were utilized to extract CT image features, followed by XGBoost, SVM, and Ensemble Voting techniques for classification, for a total of ten different classes of machine learning combinatorial classifiers. Indicators such as accuracy, sensitivity, and specificity were used to assess the models. The collected features are also shown to investigate the contrasts between them. The algorithm we presented, ResNet50-Ensemble Voting, performed best in the test set, with an accuracy of 0.943 (0.938, 0.948) and sensitivity and specificity of 0.964 and 0.911, respectively. VGG16-Ensemble Voting had an accuracy of 0.887 (0.880, 0.894), with a sensitivity and specificity of 0.952 and 0.784, respectively. Machine learning models that were implemented and integrated ResNet50-Ensemble Voting performed exceptionally well in identifying benign and malignant small pulmonary nodules (<20 mm) from various sites, which might help doctors in accurately diagnosing the nature of early-stage lung nodules in clinical practice.

Segmentation of 3d medical images for detection and classification of lung tumor using content-based features

Segmentation of 3d medical images for detection and classification of lung tumor using content-based features

CCGL-YOLOV5:A cross-modal cross-scale global-local attention YOLOV5 lung tumor detection model

BackgroundMultimodal medical image detection is a key technology in medical image analysis, which plays an important role in tumor diagnosis. There are different sizes lesions and different shapes lesions in multimodal lung tumor images, which makes it difficult to effectively extract key features of lung tumor lesions. MethodsA Cross-modal Cross-scale Clobal-Local Attention YOLOV5 Lung Tumor Detection Model (CCGL-YOLOV5) is proposed in this paper. The main works are as follows: Firstly, the Cross-Modal Fusion Transformer Module (CMFTM) is designed to improve the multimodal key lesion feature extraction ability and fusion ability through the interactive assisted fusion of multimodal features; Secondly, the Global-Local Feature Interaction Module (GLFIM) is proposed to enhance the interaction ability between multimodal global features and multimodal local features through bidirectional interactive branches. Thirdly, the Cross-Scale Attention Fusion Module (CSAFM) is designed to obtain rich multi-scale features through grouping multi-scale attention for feature fusion. ResultsThe comparison experiments with advanced networks are done. The Acc, Rec, mAP, F1 score and FPS of CCGL-YOLOV5 model on multimodal lung tumor PET/CT dataset are 97.83%, 97.39%, 96.67%, 97.61% and 98.59, respectively; The experimental results show that the performance of CCGL-YOLOV5 model in this paper are better than other typical models. ConclusionThe CCGL-YOLOV5 model can effectively use the multimodal feature information. There are important implications for multimodal medical image research and clinical disease diagnosis in CCGL-YOLOV5 model.

Optimal Thresholding for Multi-Window Computed Tomography (CT) to Predict Lung Cancer

Lung cancer is the world’s second-largest cause of cancer mortality. Patients’ lives can be saved if this malignancy is detected early. Doctors, however, encounter difficulties in detecting cancer in computed tomography (CT) images. In recent years, significant research has been devoted to producing automated lung nodule detection methods that can help radiologists. Most of them use only the lung window in their analysis and generally do not consider the mediastinal windows, which, according to recent research, carry important information. In this paper, we propose a simple yet effective algorithm to analyze multi-window CT images for lung nodules. The algorithm works in three steps. First, the CT image is preprocessed to suppress any noise and improve the image quality. Second, the lungs are extracted from the preprocessed image. Based on the histogram analysis of the lung windows, we propose a multi-Otsu-based approach for lung segmentation in lung windows. The case of mediastinal windows is rather difficult due to irregular patterns in the histograms. To this end, we propose a global–local-mean-based thresholding technique for lung detection. In the final step, the nodule candidates are extracted from the segmented lungs using simple intensity-based thresholding. The radius of the extracted objects is computed to separate the nodule from the bronchioles and blood vessels. The proposed algorithm is evaluated on the benchmark LUNA16 dataset and achieves accuracy of over 94% for lung tumor detection, surpassing that of existing similar methods.

Pathological Characteristics of Fiberoptic Bronchoscopy Procedure in Ulin General Hospital, Banjarmasin

Introduction: Fiberoptic Bronchoscopy (FOB) has emerged into the method of choice for the detection of lung tumors due to its availability of Adequate cytological and histological samples in the form of broncho-alveolar lavage, bronchial brushings, and bronchial forceps biopsy. This study aims to identify characteristics of bronchoalveolar lavage, aspiration cytology, brushing cytology and forceps cytology of patients suspected with malignancy at the Ulin Regional Hospital in South Kalimantan. Methods: A cross-sectional study was carried out utilizing medical documentation of patients suspested lung cancer from March 2018 to March 2020. Data were grouped by gender, age, type of procedure, and cytology and histopathology findings. Results: The result showed that 67.5% of 117 patients who underwent bronchoscopy had positive results and common histological diagnosis The frequent histopathological conclusion was adenocarcinoma (45.6%) tailed by squamous cell carcinoma (40.55). Forceps bronchoscopy showed a greater positivity rate (75.0%) than bronchoscopy without forceps (63.6%), we also found that by combining BAL and brushing and forceps biopsy, the detection rate of lung malignancy increased to 85.7%, but no correlation between the type of procedure and the likelihood of malignancy was discovered. Conclusions: Forceps bronchoscopy showed a greater positivity rate than bronchoscopy without forceps, but no correlation between the type of procedure and the likelihood of malignancy was discovered in our study.

Radiolabeled gemcitabine hydrochloride as an imaging agent for lung cancer: Radiolabeling, quality control and cell incorporation studies

The development of new drugs that can specifically screen tumors is a global need. When it comes to lung cancer, which is the second main cause of cancer-related deaths, early detection of lung tumors using appropriate imaging is very important. In this study, gemcitabine hydrochloride (GCH) was radiolabeled with [99mTc]Tc under different conditions (changing reducing agent, antioxidant agent, incubation time, pH, [99mTc]Tc activity) and radiolabeling activity (quality control) using Radio Thin Layer Chromatography and paper electrophoresis. The results showed that the most stable complex ([99mTc]Tc-GCH) was prepared using 0.015 mg of stannous chloride as a reducing agent, 0.01 mg of ascorbic acid as an antioxidant and 37 MBq activity at pH 7.4 after 15 min of incubation time. The complex remained stable for 6 h. Cell incorporation studies showed a six-fold higher uptake of [99mTc]Tc-GCH in cancer (A-549) cells (38.42 ± 1.53) than healthy (L-929) cells (6.11 ± 0.17) have shown that it can. In addition, the different behaviors of R/H-[99mTc]Tc confirmed the specificity of this newly developed radiopharmaceutical. Although these studies are preliminary, it has been concluded that [99mTc]Tc-GCH may be a candidate drug for use in nuclear medicine, particularly in the diagnosis of lung cancer.

ELCT-YOLO: An Efficient One-Stage Model for Automatic Lung Tumor Detection Based on CT Images

Research on lung cancer automatic detection using deep learning algorithms has achieved good results but, due to the complexity of tumor edge features and possible changes in tumor positions, it is still a great challenge to diagnose patients with lung tumors based on computed tomography (CT) images. In order to solve the problem of scales and meet the requirements of real-time detection, an efficient one-stage model for automatic lung tumor detection in CT Images, called ELCT-YOLO, is presented in this paper. Instead of deepening the backbone or relying on a complex feature fusion network, ELCT-YOLO uses a specially designed neck structure, which is suitable to enhance the multi-scale representation ability of the entire feature layer. At the same time, in order to solve the problem of lacking a receptive field after decoupling, the proposed model uses a novel Cascaded Refinement Scheme (CRS), composed of two different types of receptive field enhancement modules (RFEMs), which enables expanding the effective receptive field and aggregate multi-scale context information, thus improving the tumor detection performance of the model. The experimental results show that the proposed ELCT-YOLO model has strong ability in expressing multi-scale information and good robustness in detecting lung tumors of various sizes.

Indocyanine green inhalation visualizes lung tumour during video-assisted thoracoscopic surgery.

Accurate intraoperative identification of small lung tumours is crucial for precise resection of these lesions during video-assisted thoracoscopic surgery. This study aimed to evaluate the feasibility and safety of indocyanine green (ICG) inhalation for intraoperative visualization of lung tumours. From January 2022 to May 2022, 43 patients with lung nodules were included into this study. All patients received intraoperative ICG inhalation for visualization of lung tumours under near-infrared imaging. The primary outcomes of this trial were the detection rate and background-tumour ratio of lung nodules, and the secondary objectives were time to search for nodules and operative time to nodules excision. A total of 50 pulmonary nodules in 43 patients were identified and completely resected. And 44 lung nodules were detected during intraoperative fluorescent exploration with a median inhaled ICG dose of 18.8 mg. In vivo, the median background-tumour ratio was 7.10. The median detection time of nodules was 100 s and the median operative time to nodules excision was 18 min. Quantification analysis showed that the fluorescence intensity of postoperative sputum declined to ∼10% of the first fluorescent sputum within 20 h. No adverse events attributed to ICG inhalation were recorded during the follow-up period. Intraoperative inhalation of ICG was a feasible and safe method for detection of lung tumours at low dose of ICG. This technique could be a remedial measure for identification of unpalpable lung nodules without preoperative localization. Chinese Clinical Trial Registry, Identifier: ChiCTR2100053708.