Early Diagnosis Of Lung Cancer Research Articles

S-Net: an S-shaped network for nodule detection in 3D CT images

Objective. Accurate and automatic detection of pulmonary nodules is critical for early lung cancer diagnosis, and promising progress has been achieved in developing effective deep models for nodule detection. However, most existing nodule detection methods merely focus on integrating elaborately designed feature extraction modules into the backbone of the detection network to extract rich nodule features while ignore disadvantages of the structure of detection network itself. This study aims to address these disadvantages and develop a deep learning-based algorithm for pulmonary nodule detection to improve the accuracy of early lung cancer diagnosis. Approach. In this paper, an S-shaped network called S-Net is developed with the U-shaped network as backbone, where an information fusion branch is used to propagate lower-level details and positional information critical for nodule detection to higher-level feature maps, head shared scale adaptive detection strategy is utilized to capture information from different scales for better detecting nodules with different shapes and sizes and the feature decoupling detection head is used to allow the classification and regression branches to focus on the information required for their respective tasks. A hybrid loss function is utilized to fully exploit the interplay between the classification and regression branches. Main results. The proposed S-Net network with ResSENet and other three U-shaped backbones from SANet, OSAF-YOLOv3 and MSANet (R+SC+ECA) models achieve average CPM scores of 0.914, 0.915, 0.917 and 0.923 on the LUNA16 dataset, which are significantly higher than those achieved with other existing state-of-the-art models. Significance. The experimental results demonstrate that our proposed method effectively improves nodule detection performance, which implies potential applications of the proposed method in clinical practice.

BiCFormer: Swin Transformer based model for classification of benign and malignant pulmonary nodules

Pulmonary cancer is one of the most common and deadliest cancers worldwide, and the detection of benign and malignant nodules in the lungs can be an important aid in the early diagnosis of lung cancer. Existing convolutional neural networks inherit their limitations by extracting global contextual information, and in most cases prove to be less efficient in obtaining satisfactory results. Transformer-based deep learning methods have obtained good performance in different computer vision tasks, and this study attempts to introduce them into the task of computed tomography (CT) image classification of lung nodules. However, the problems of sample scarcity and difficulty of local feature extraction in this field. To this end, we are inspired by Swin Transformer to propose a model named BiCFormer for the task of classifying and diagnosing CT scan images of lung nodules. Specifically, first we introduce a multi-layer discriminator generative adversarial network module for data augmentation to assist the model in extracting features more accurately. Second, unlike the encoder of traditional Transformer, we divide the encoder part of BiCFormer into two parts: bi-level coordinate (BiC) and fast-partial-window (FPW). The BiC module has a part similar to the traditional channel attention mechanism is able to enhance the performance of the model, and is more able to enhance the representation of attention object features by aggregating features along two spatial directions. The BiC module also has a dynamic sparse attention mechanism that filters out irrelevant key-value pairs in rough regions, allowing the model to focus more on features of interest. The FPW module is mainly used to reduce computational redundancy and minimize feature loss. We conducted extensive experiments on the LIDC-IDRI dataset. The experimental results show that our model achieves an accuracy of 97.4% compared to other studies using this dataset for lung nodule classification, making it an effective and competitive method.

Circulating microbiome DNA as biomarkers for early diagnosis and recurrence of lung cancer

Lung cancer mortality is exacerbated by late-stage diagnosis. Emerging evidence indicates the potential clinical significance of distinct microbial signatures as diagnostic and prognostic biomarkers across various cancers. However, circulating microbiome DNA (cmDNA) profiles are underexplored in lung cancer (LC). Here, whole-genome sequencing is performed on plasma of LC patients and healthy controls (HCs). Differentially enriched microbial species are identified between LC and HC. A diagnostic model is developed, which has a high sensitivity of 87.7% and achieves an AUC of 93.2% in the independent validation dataset. Crucially, this model demonstrates the capability to detect early-stage LC, achieving a sensitivity of 86.5% for stage I and 87.1% for tumors <1cm. In addition, we construct a cmDNA model for recurrence, which precisely predicts LC recurrence after surgery. Overall, this study highlights the significant alterations of cmDNA profiles in LC, indicating its potential as biomarkers for early diagnosis and recurrence.

Tandem hybridization chain reaction and selective coordination enable fluorescence detection of exosomes in lung cancer

Liquid biopsy based on tumor-derived exosomes (TDEs) is potentially effective in assisting early cancer diagnosis. However, accurate quantification of the approach remains challenging due to the low concentration and heterogeneity of TDEs. This work involved designing a homogeneous enzyme-free fluorescent sensing strategy by targeting programmed cell death ligand 1 (PD-L1) that was overexpressed on the surface of TDEs, coupled with the tandem hybridization chain reaction (HCR) and the selective coordination of Cu2+. The aptamer was used as a specific recognition probe for binding with the target protein, thereby regulating the tandem HCR reactions and achieving signal amplification. The coordination between Cu2+ and double-stranded DNA (dsDNA) can further regulate the fluorescence intensity of the signal molecule, calcein, for precise quantification of TDEs. The results showed that the detection limit (LOD) for TDEs can be as low as 100 particles/mL. Moreover, the proposed strategy can accurately differentiate between lung cancer patients (n = 25) and healthy donors (n = 12), with a sensitivity of 92%, specificity of 100%, and an area under the receiver operating characteristic curve (AUC) of 0.963. The fluorescence analysis also demonstrated a strong correlation with the computed tomography scans and pathological analysis. Therefore, the proposed method in this study can provide more insights into assisting in the early diagnosis of lung cancer through TDEs-based liquid biopsy.

Lung cancer after kidney transplantation: a 50-year experience at a single institution.

To investigate the clinical characteristics of lung cancer that develops after kidney transplantation. The clinical data of patients with lung cancer diagnosed after kidney transplantation were collected retrospectively. The medical records were extracted from our database. All patients underwent routine chest examination after kidney transplantation. In total, 17 lung tumors were detected in 15 (0.6%) of 2593 patients who underwent kidney transplantation at our institution. Eleven lung tumors were completely resected from a collective 10 patients (surgical group). The remaining five patients did not receive surgical treatment (nonsurgical group). The surgical group underwent wedge resection (n = 5), segmentectomy (n = 1), lobectomy (n = 3), and bilobectomy (n = 1). The pathological stages were 0 (n = 1), IA1 (n = 2), IA2 (n = 4), IA3 (n = 2), and IB (n = 1). The surgical group had a significantly better prognosis than the nonsurgical group. There were no perioperative complications related to kidney transplantation in either group. Routine chest examination would be useful for the early diagnosis and treatment of lung cancer after kidney transplantation. Moreover, surgical resection for early-stage lung cancer was associated with a better prognosis for kidney transplantation patients.

Deep Reinforcement Learning for Early Diagnosis of Lung Cancer

Lung cancer remains the leading cause of cancer-related death worldwide, and early diagnosis of lung cancer is critical for improving the survival rate of patients. Performing annual low-dose computed tomography (LDCT) screening among high-risk populations is the primary approach for early diagnosis. However, after each screening, whether to continue monitoring (with follow-up screenings) or to order a biopsy for diagnosis remains a challenging decision to make. Continuing with follow-up screenings may lead to delayed diagnosis but ordering a biopsy without sufficient evidence incurs unnecessary risk and cost. In this paper, we tackle the problem by an optimal stopping approach. Our proposed algorithm, called EarlyStop-RL, utilizes the structure of the Snell envelope for optimal stopping, and model-free deep reinforcement learning for making diagnosis decisions. Through evaluating our algorithm on a commonly used clinical trial dataset (the National Lung Screening Trial), we demonstrate that EarlyStop-RL has the potential to greatly enhance risk assessment and early diagnosis of lung cancer, surpassing the performance of two widely adopted clinical models, namely the Lung-RADS and the Brock model.

Utilizing Magnetic Levitation to Detect Lung Cancer-Associated Exosomes.

Extracellular vesicles, especially exosomes, have attracted attention in the last few decades as novel cancer biomarkers. Exosomal membrane proteins provide easy-to-reach targets and can be utilized as information sources of their parent cells. In this study, a MagLev-based, highly sensitive, and versatile biosensor platform for detecting minor differences in the density of suspended objects is proposed for exosome detection. The developed platform utilizes antibody-functionalized microspheres to capture exosomal membrane proteins (ExoMPs) EpCAM, CD81, and CD151 as markers for cancerous exosomes, exosomes, and non-small cell lung cancer (NSCLC)-derived exosomes, respectively. Initially, the platform was utilized for protein detection and quantification by targeting solubilized ExoMPs, and a dynamic range of 1-100 nM, with LoD values of 1.324, 0.638, and 0.722 nM for EpCAM, CD81, and CD151, were observed, respectively. Then, the sensor platform was tested using exosome isolates derived from NSCLC cell line A549 and MRC5 healthy lung fibroblast cell line. It was shown that the sensor platform is able to detect and differentiate exosomal biomarkers derived from cancerous and non-cancerous cell lines. Overall, this innovative, simple, and rapid method shows great potential for the early diagnosis of lung cancer through exosomal biomarker detection.

Abstract 6090: Integration of innate and adaptive immune signatures for early detection of cancer

Abstract Up to 50% of cancer patients are diagnosed at a late stage with tumors that are often unresectable, leading to intensive treatments and a preventable loss of life. Whilst multiple assays have been developed to tackle the issue of disease mortality, these approaches have serious limitations. For example, circulating tumor DNA (ctDNA) preferentially detects hard-to-treat tumors and micrometastatic disease, with limited sensitivity and specificity for early-stage malignancy, underscoring an urgent need for novel early detection strategies. Substantial evidence exists to demonstrate that signals related to innate (e.g. pro-inflammatory cytokines) and adaptive (e.g. antibodies and T cell receptors [TCRs]) immune engagement arise early in the development of cancer. The immune system therefore offers potential as an exquisitely sensitive and highly specific intrinsic early detection system for cancer. Here we propose the development of a pan-cancer novel early detection assay, using peripheral blood to measure three key cancer-specific components of the anti-tumor immune response: i) T cell receptor (TCR) sequences, ii) antibody signatures, iii) cytokine markers. Our preliminary data demonstrates that an age and smoking-matched cohort of lung cancer patients can be distinguished from healthy donors using TCR sequencing data processed through a published machine learning approach (p&amp;lt;0.01). In support of this, a pilot study on a custom peptide microarray comprised of lung cancer associated antigens and recurrent neoantigens demonstrated that lung cancer patient samples show significantly higher antibody intensities than healthy donors. Thirdly, using data from a prospective cohort study, we have demonstrated that inflammatory markers can be leveraged to predict future cancer diagnosis, and identified key protein markers for future work. These data provide proof-of-principle for the use of these analytes as a tool for early detection. We are actively curating a unique cohort of samples for this work from a prospective study, Nodule Immunophenotyping Biomarker for Lung Cancer Early Diagnosis (NIMBLE) (NCT05432739), which recruits patients with indeterminate lung nodules that may represent early cancers (&amp;gt;280 patients to date). This malignancy has high mortality rates and a significant need for early detection strategies. Through this, we aim to understand the early immunobiological response to cancer and leverage this information to design a novel multi-parametric immunopredictor assay for early detection of cancer. Citation Format: Evelyn Fitzsimons, Alexander Coulton, Hongchang Fu, Marcellus Augustine, Richard Lee, James Reading, Kevin Litchfield. Integration of innate and adaptive immune signatures for early detection of cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6090.

Abstract 4151: Development of a cancer-activated biologic imaging platform for early lung cancer diagnosis

Abstract Current modalities, such as LDCT and fluorodeoxyglucose positron emission tomography (FDG PET), for early lung cancer imaging, rely upon proxies resulting in poor specificity and/or sensitivity. They are also unreliable for discriminating benign versus malignant lesions resulting in the requirement of longitudinal analyses euphemistically called “watchful waiting”. Lung biopsies are marred by significant false negative rates and pose complications of collapsed lungs or bleeding in up to 22% of procedures and death rates up to 1%. To overcome these obstacles, Earli is developing a non-invasive, highly specific cancer-activated imaging platform that usurps dysregulated gene expression within malignancies to force only cancer cells to produce a synthetic biomarker that can be localized using PET imaging. EARLI-204 is a lipid nanoparticle that contains a DNA nanoplasmid comprised of a cancer-activated promoter driving selective expression of a functionally inactivated somatostatin receptor 2 (SSTR2fi) gene. Abrogation of internalization and signal transduction allows the safe use of SSTR2fi as an ectopically expressed PET reporter gene (PRG). Following administration, EARLI-204 transfects tumor and normal cells but transcriptional activation only within cancer cells results in selective cell-surface expression of SSTR2fi that can be imaged with FDA-approved SSTR2 PET tracers. H1299 human lung cancer cells engineered to express SSTR2 (H1299-SSTR2) revealed a discernable PET signal with as little as 31,000 engineered cells. Mixed cell titration studies using subcutaneous (SQ) tumors with H1299-SSTR2 and non-expressing H1299-wt cells suggest tumors with between 0.1% and 1% H1299-SSTR2 cells had a significant PET signal, compared to tumors with 100% H1299-wt. This result sets a benchmark transfection of ~1% to achieve a PET signal. Intratumoral (IT) dose titration of EARLI-204 revealed that 5 ug of EARLI-204 generates a significant PET signal. Consistent with the H1299-SSTR2 studies, IHC shows ~1% of tumor cells express the PRG at a 5 ug IT dose level. Furthermore, we measured 70 copies of EARLI-204 DNA per cell resulting in 22,629 SSTR2 mRNA copies/25 ng at a 5 ug IT dose level. IV dosing of EARLI-204 in SQ and intralung tumor models results in cancer-activated expression of SSTR2fi in both SQ and small (&amp;lt;4 mm) lung tumors with significantly increased (p &amp;lt; 0.05) PET signal. For lung tumors, analyses showed 1 copy of EARLI-204 DNA/cell resulting in 780 SSTR2 mRNA copies/25 ng. The results demonstrate an effective molecular imaging platform for drug discovery ranging from in vivo screening to building PK-PD-efficacy relationships and enabling human dose projection for the Earli PET localization drug. The benchmarks and findings established with this preclinical imaging pipeline are currently being evaluated in large animals such as dogs and pigs, and will be translated to humans in planned clinical trials. Citation Format: Mohammed Goryawala, Hung-Yu Henry Lee, Ling Tong, Trupti Patil, Alex Harwig, Chloe Xia, Suthara Ramachandran, Dariusz Wodziak, Dean Felsher, Tim Sproul, David Suhy. Development of a cancer-activated biologic imaging platform for early lung cancer diagnosis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4151.

Novel Image-Guided Percutaneous Lung Tissue Excision Device With Integrated Sealing of Blood Vessels and Airways: An In Vivo Preclinical Study.

This study evaluated the efficacy of the Minimally Invasive Targeted Resection (MiTR) device, a novel electrosurgical instrument that allows for targeted excision of a lung abnormality while using bipolar radiofrequency (RF) energy to seal blood vessels and airways. The MiTR system was evaluated in 7 acute and 2 chronic porcine (7-day) models to evaluate the efficacy of tissue excision with bipolar RF sealing of blood vessels and airways and application of an autologous blood patch into the excised tissue cavity. Air leak was recorded for all evaluations. The study was approved by the institutional ethical board. Nineteen lung tissue samples, measuring 2.5 cm long × 1.2 cm diameter, were excised. In 8 of 9 animals (89%), hemostasis and pneumostasis were observed visually at the completion of the procedure. In 2 of 2 chronic animals (100%), hemostasis and pneumostasis persisted for the 7-day observation period. Histologic examination of the excised samples showed preservation of the core parenchymal architecture without evident tissue damage of the samples that would impair pathologic analysis. Percutaneous resection of targeted lung tissue with the MiTR system demonstrated hemostasis and pneumostasis while obtaining a histologically intact sample. After regulatory approval, the use of this device could offer more tissue for analysis than a transthoracic needle biopsy or bronchoscopy and a far less invasive alternative to video-assisted thoracic surgery or thoracotomy. This may also expand patient and physician options for the early diagnosis and treatment of lung cancer.

Ru-Doped MoS2 Monolayer for Exhaled Breath Detection on Early Lung Cancer Diagnosis: A First-Principles Investigation.

Nanosensor-based patient exhaled breath detection is a practical and effective way to detect lung cancer early. In this paper, a Ru-doped MoS2 monolayer (Ru-MoS2) is proposed as a promising novel biosensor based on first-principles theory for the detection of three typical early stage lung cancer exhaled volatile organic compounds, namely, C3H4O, C3H6O, and C5H8. Replacement of a S atom in the MoS2 monolayer with a Ru dopant atom to form a stable Ru-MoS2 monolayer with a binding energy of -4.78 eV is further demonstrated by the thermostability and chemical stability analysis as well as improving the adsorption performance of the system for three VOCs. The adsorption configuration structures, adsorption properties, and electronic behavior of the Ru-MoS2 monolayer are investigated by electron deformation density and density of states analysis to gain a comprehensive understanding of the physicochemical properties as sensing material. The results show that the adsorption energies of the Ru-MoS2 monolayer for C3H4O, C3H6O, and C5H8 are 3.42, -1.53, and -2.80 eV, respectively, all of which are chemisorption with excellent adsorption performance. The sensitivities for the three VOCs could be up to 1.09, 140.50, and 5.90, respectively, and the band structure and work function further elucidate the sensing mechanism of the Ru-MoS2 monolayer as a resistive gas sensor. The type and concentration of these exhaled breaths may reflect changes in the patient's physiological and biochemical status and may serve as a probe for the diagnosis of lung cancer. The results in this work could provide a guidance for researchers to explore the practical applications in the early diagnosis of lung cancer by gas sensors.

Dual-branch feature fusion S3D V-Net network for lung nodules segmentation.

Accurate segmentation of lung nodules can help doctors get more accurate results and protocols in early lung cancer diagnosis and treatment planning, so that patients can be better detected and treated at an early stage, and the mortality rate of lung cancer can be reduced. Currently, the improvement of lung nodule segmentation accuracy has been limited by his heterogeneous performance in the lungs, the imbalance between segmentation targets and background pixels, and other factors. We propose a new 2.5D lung nodule segmentation network model for lung nodule segmentation. This network model can well improve the extraction of edge information of lung nodules, and fuses intra-slice and inter-slice features, which makes good use of the three-dimensional structural information of lung nodules and can more effectively improve the accuracy of lung nodule segmentation. Our approach is based on a typical encoding-decoding network structure for improvement. The improved model captures the features of multiple nodules in both 3-D and 2-D CT images, complements the information of the segmentation target's features and enhances the texture features at the edges of the pulmonary nodules through the dual-branch feature fusion module (DFFM) and the reverse attention context module (RACM), and employs central pooling instead of the maximal pooling operation, which is used to preserve the features around the target and to eliminate the edge-irrelevant features, to further improve the performance of the segmentation of the pulmonary nodules. We evaluated this method on a wide range of 1186 nodules from the LUNA16 dataset, and averaging the results of ten cross-validated, the proposed method achieved the mean dice similarity coefficient (mDSC) of 84.57%, the mean overlapping error (mOE) of 18.73% and average processing of a case is about 2.07 s. Moreover, our results were compared with inter-radiologist agreement on the LUNA16 dataset, and the average difference was 0.74%. The experimental results show that our method improves the accuracy of pulmonary nodules segmentation and also takes less time than more 3-D segmentation methods in terms of time.

Transcriptome profiling and metabolic pathway analysis towards reliable biomarker discovery in early-stage lung cancer.

Earlier diagnosis of lung cancer is crucial for reducing mortality and morbidity in high-risk patients. Liquid biopsy is a critical technique for detecting the cancer earlier and tracking the treatment outcomes. However, noninvasive biomarkers are desperately needed due to the lack of therapeutic sensitivity and early-stage diagnosis. Therefore, we have utilized transcriptomic profiling of early-stage lung cancer patients to discover promising biomarkers and their associated metabolic functions. Initially, PCA highlights the diversity level of gene expression in three stages of lung cancer samples. We have identified two major clusters consisting of highly variant genes among the three stages. Further, a total of 7742, 6611, and 643 genes were identified as DGE for stages I-IIIrespectively. Topological analysis of the protein-protein interaction network resulted in seven candidate biomarkers such as JUN, LYN, PTK2, UBC, HSP90AA1, TP53, and UBB cumulatively for the three stages of lung cancers. Gene enrichment and KEGG pathway analyses aid in the comprehension of pathway mechanisms and regulation of identified hub genes in lung cancer. Importantly, the medial survival rates up to ~ 70months were identified for hub genes during the Kaplan-Meier survival analysis. Moreover, the hub genes displayed the significance of risk factors during gene expression analysis using TIMER2.0 analysis. Therefore, we have reason that these biomarkers may serve as a prospective targeting candidate with higher treatment efficacy in early-stage lung cancer patients.

Breath analysis system with convolutional neural network (CNN) for early detection of lung cancer

Early diagnosis of lung cancer, the leading cause of cancer-related death worldwide, is critical for reducing mortality rate. However, current diagnostic methods are invasive, time-consuming, costly, and may not always provide accurate diagnoses. For early diagnosis, recent research has focused on noninvasive approaches, including the detection of volatile organic compounds (VOCs) in human exhaled breath. Changes in the composition and concentration of VOCs in exhaled breath may indicate lung cancer, and this approach offers several advantages over traditional diagnostic methods. Moreover, the combination of a breath gas sensing system and machine learning algorithms provides a more accurate diagnosis. In this study, for the early diagnosis of lung cancer, a breath analysis system was developed using a gas sensor array and deep learning algorithm. The breath analysis system was designed to detect multiple VOCs in exhaled breath using ten semiconductor metal oxide (SMO), one photoionization detector (PID), nine electrochemical (EC) gas sensors. In total, 181 clinical breath samples (from 74 healthy controls and 107 lung cancer patients) were collected and analyzed using a 1D convolutional neural network (CNN) algorithm. The results showed an overall accuracy of 97.8% in classifying healthy controls and lung cancer patients using a complete clinical dataset. Through a comparison of the single-sensor type data and multimodal sensor data and performance analysis of three different deep learning models (multilayer perceptron, recurrent neural network, and CNN), we validated the potential of the breath analyzer with a multimodal sensor system and a 1D CNN as a lung cancer diagnostic device.

Recommended approaches for screening and early detection of lung cancer in the Middle East and Africa (MEA) region: a consensus statement.

The prevalence of lung cancer in the Middle East and Africa (MEA) region has steadily increased in recent years and is generally associated with a poor prognosis due to the late detection of most of the cases. We explored the factors leading to delayed diagnoses, as well as the challenges and gaps in the early screening, detection, and referral framework for lung cancer in the MEA. A steering committee meeting was convened in October 2022, attended by a panel of ten key external experts in the field of oncology from the Kingdom of Saudi Arabia, United Arab Emirates, South Africa, Egypt, Lebanon, Jordan, and Turkey, who critically and extensively analyzed the current unmet needs and challenges in the screening and early diagnosis of lung cancer in the region. As per the experts' opinion, lack of awareness about disease symptoms, misdiagnosis, limited screening initiatives, and late referral to specialists were the primary reasons for delayed diagnoses emphasizing the need for national-level lung cancer screening programs in the MEA region. Screening guidelines recommend low-dose computerized tomography (LDCT) for lung cancer screening in patients with a high risk of malignancy. However, high cost and lack of awareness among the public as well as healthcare providers prevented the judicious use of LDCT in the MEA region. Well-established screening and referral guidelines were available in only a few of the MEA countries and needed to be implemented in others to identify suspected cases early and provide timely intervention thus improving patient outcomes. There is a great need for large-scale screening programs, preferably integrated with tobacco-control programs and awareness programs for physicians and patients, which may facilitate higher adherence to lung cancer screening and improve survival outcomes.

Establishment and verification of a prediction model based on clinical characteristics and computed tomography radiomics parameters for distinguishing benign and malignant pulmonary nodules.

The radiographic classification of pulmonary nodules into benign versus malignant categories is a pivotal component of early lung cancer diagnosis. The present study aimed to investigate clinical and computed tomography (CT) clinical-radiomics nomogram for preoperative differentiation of benign and malignant pulmonary nodules. This retrospective study included 342 patients with pulmonary nodules who underwent high-resolution CT (HRCT) examination. We assigned them to a training dataset (n=239) and a validation dataset (n=103). There are 1781 tumor characteristics quantified by extracted features from the lesion segmented from patients' CT images. The features with poor reproducibility and high redundancy were removed. Then a least absolute shrinkage and selection operator (LASSO) logistic regression model with 10-fold cross-validation was used to further select features and build radiomics signatures. The independent predictive factors were identified by multivariate logistic regression. A radiomics nomogram was developed to predict the malignant probability. The performance and clinical utility of the clinical-radiomics nomogram was evaluated by receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). After dimension reduction by the LASSO algorithm and multivariate logistic regression, four radiomic features were selected, including original_shape_Sphericity, exponential_glcm_Maximum Probability, log_sigma_2_0_mm_3D_glcm_Maximum Probability, and ogarithm_firstorder_90Percentile. Multivariate logistic regression showed that carcinoembryonic antigen (CEA) [odds ratio (OR) 95% confidence interval (CI): 1.40 (1.09-1.88)], CT rad score [OR (95% CI): 2.74 (2.03-3.85)], and cytokeratin-19-fragment (CYFRA21-1) [OR (95% CI): 1.80 (1.14-2.94)] were independent influencing factors of malignant pulmonary nodule (all P<0.05). The clinical-radiomics nomogram combining CEA, CYFRA21-1 and radiomics features achieved an area of curve (AUC) of 0.85 and 0.76 in the training group and verification group for the prediction of malignant pulmonary nodules. The clinical-radiomics nomogram demonstrated excellent agreement and practicality, as evidenced by the calibration curve and DCA. The clinical-radiomics nomogram combined of CT-based radiomics signature, along with CYFRA21-1 and CEA, demonstrated strong predictive ability, calibration, and clinical usefulness in distinguishing between benign and malignant pulmonary nodules. The use of CT-based radiomics has the potential to assist clinicians in making informed decisions prior to biopsy or surgery while avoiding unnecessary treatment for non-cancerous lesions.

METTL3-mediated m6A modification enhances ZDHHC16 expression in nonsmall- cell lung cancer patients, attenuating ferroptosis by suppressing CREB ubiquitination.

At present, the early diagnosis and treatment of non-small cell lung cancer (NSCLC) is still an urgent problem to be solved worldwide, including in China. The present work investigated the possible protective effect of ZDHHC16 in cell proliferation and metastasis of NSCLC and explored its possible mechanisms. ZDHHC16 expression level in patients with Non-Small-Cell Lung Cancer was up-regulation. ZDHHC16 gene is stabilized by m6A methylation. ZDHHC16 gene reduced ferroptosis of NSCLC by the rehabilitation of the mitochondrial structure. ZDHHC16 promoted CREB expression through the inhibition of CREB Ubiquitination. Confocalmicroscopy showed that ZDHHC16 reduced the CREB expression of NSCLC. ZDHHC16 up-regulation reduced CREB Ubiquitination, and down-regulation of ZDHHC16 promoted CREB Ubiquitination of NSCLC. CREB Agonists reduced the effects of ZDHHC16 on ferroptosis, not affecting the Warburg effect of NSCLC. CREB inhibitor reduced the effects of si-ZDHHC16 on ferroptosis, not affecting the Warburg effect of NSCLC. METTL3-mediated m6A modification increases ZDHHC16 stability. Our study revealed that the m6A-forming enzyme METTL3 upregulates ZDHHC16 expression in NSCLC patients, leading to the reduction of ferroptosis by inhibiting CREB ubiquitination.

An improved V-Net lung nodule segmentation model based on pixel threshold separation and attention mechanism.

Accurate labeling of lung nodules in computed tomography (CT) images is crucial in early lung cancer diagnosis and before nodule resection surgery. However, the irregular shape of lung nodules in CT images and the complex lung environment make it much more challenging to segment lung nodules accurately. On this basis, we propose an improved V-Net segmentation method based on pixel threshold separation and attention mechanism for lung nodules. This method first offers a data augment strategy to solve the problem of insufficient samples in 3D medical datasets. In addition, we integrate the feature extraction module based on pixel threshold separation into the model to enhance the feature extraction ability under different thresholds on the one hand. On the other hand, the model introduces channel and spatial attention modules to make the model pay more attention to important semantic information and improve its generalization ability and accuracy. Experiments show that the Dice similarity coefficients of the improved model on the public datasets LUNA16 and LNDb are 94.9% and 81.1% respectively, and the sensitivities reach 92.7% and 76.9% respectively. which is superior to most existing UNet architecture models and comparable to the manual level segmentation results by medical technologists.

Overcoming the Challenge of Accurate Segmentation of Lung Nodules: A Multi-crop CNN Approach.

Lung nodules are generated based on the growth of small and round- or oval-shaped cells in the lung, which are either cancerous or non-cancerous. Accurate segmentation of these nodules is crucial for early detection and diagnosis of lung cancer. However, lung nodules can have various shapes, sizes, and densities, making their accurate segmentation a difficult task. Moreover, they can be easily confused with other structures in the lung, including blood vessels and airways, further complicating the segmentation process. To address this challenge, this paper proposes a novel multi-crop convolutional neural network (multi-crop CNN) model that utilizes different sized cropped regions of CT scan images for accurate segmentation of lung nodules. The model consists of three modules, namely the feature representation module, boundary refinement module, and segmentation module. The feature representation module captures features from the lung CT scan image using cropped regions of different sizes, while the boundary refinement module combines the boundary maps and feature maps to generate a final feature map for the segmentation process. The segmentation module produces a high-resolution segmentation map that shows improved accuracy in segmenting cancerous lung nodules. The proposed multi-crop CNN model is evaluated on two segmentation datasets namely LUNA 16 and LIDC-IDRI with an accuracy of 98.3% and 98.5%, respectively. The performances are measured in terms of accuracy, recall, precision, dice coefficient, specificity, AUC/ROC, Hausdorff distance, Jaccard index, and average Hausdorff. Overall, the proposed multi-crop CNN model demonstrates the potential to enhance the lung nodule segmentation accuracy, which could lead to earlier detection and diagnosis of lung cancer and ultimately reduce mortality rates associated with the disease.

Urinary VOCs as biomarkers of early stage lung tumour development in mice.

Lung cancer is the primary cause of cancer-induced death. In addition to prevention and improved treatment, it has increasingly been established that early detection is critical to successful remission. The aim of this study was to identify volatile organic compounds (VOCs) in urine that could help diagnose mouse lung cancer at an early stage of its development. We analysed the VOC composition of urine in a genetically engineered lung adenocarcinoma mouse model with oncogenic EGFR doxycycline-inducible lung-specific expression. We compared the urinary VOCs of 10 cancerous mice and 10 healthy mice (controls) before and after doxycycline induction, every two weeks for 12 weeks, until full-blown carcinomas appeared. We used SPME fibres and gas chromatography- mass spectrometry to detect variations in cancer-related urinary VOCs over time. This study allowed us to identify eight diagnostic biomarkers that help discriminate early stages of cancer tumour development (i.e., before MRI imaging techniques could identify it). The analysis of mice urinary VOCs have shown that cancer can induce changes in odour profiles at an early stage of cancer development, opening a promising avenue for early diagnosis of lung cancer in other models.