Lung Nodules Research Articles

Airborne Aluminum as an Underestimated Source of Human Exposure: Quantification of Aluminum in 24 Human Tissue Types Reveals High Aluminum Concentrations in Lung and Hilar Lymph Node Tissues.

Aluminum (Al) is the most abundant metal in the earth's crust, and humans are exposed to Al through sources like food, cosmetics, and medication. So far, no comprehensive data on the Al distribution between and within human tissues were reported. We measured Al concentrations in 24 different tissue types of 8 autopsied patients using ICP-MS/MS (inductively coupled plasma-tandem mass spectrometry) under cleanroom conditions and found surprisingly high concentrations in both the upper and inferior lobes of the lung and hilar lymph nodes. Al/Si ratios in lung and hilar lymph node samples of 12 additional patients were similar to the ratios reported in urban fine dust. Histological analyses using lumogallion staining showed Al in lung erythrocytes and macrophages, indicating the uptake of airborne Al in the bloodstream. Furthermore, Al was continuously found in PM2.5 and PM10 fine dust particles over 7 years in Upper Austria, Austria. According to our findings, air pollution needs to be reconsidered as a major Al source for humans and the environment.

Glycan Profiling in Small Extracellular Vesicles with a SERS Microfluidic Biosensor Identifies Early Malignant Development in Lung Cancer.

Glycosylation is the most common post-translational modification of proteins and regulates a myriad of fundamental biological processes under normal, and pathological conditions. Altered protein glycosylation is linked to malignant transformation, showing distinct glycopatterns that are associated with cancer initiation and progression by regulating tumor proliferation, invasion, metastasis, and therapeutic resistance. The glycopatterns of small extracellular vesicles (sEVs) released by cancer cells are promising candidates for cancer monitoring since they exhibit glycopatterns similar to their cell-of-origin. However, the clinical application of sEV glycans is challenging due to the limitations of current analytical technologies in tracking the trace amounts of sEVs specifically derived from tumors in circulation. Herein, a sEV GLYcan PHenotype (EV-GLYPH) assay that utilizes a microfluidic platform integrated with surface-enhanced Raman scattering for multiplex profiling of sEV glycans in non-small cell lung cancer is clinically validated. For the first time, the EV-GLYPH assay effectively identifies distinct sEV glycan signatures between non-transformed and malignantly transformed lung cells. In a clinical study evaluated on 40 patients, the EV-GLYPH assay successfully differentiates patients with early-stage malignant lung nodules from benign lung nodules. These results reveal the potential to profile sEV glycans for noninvasive diagnostics and prognostics, opening up promising avenues for clinical applications and understanding the role of sEV glycosylation in lung cancer.

Optimal approach for diagnosing peripheral lung nodules by combining electromagnetic navigation bronchoscopy and radial probe endobronchial ultrasound.

Electromagnetic navigation bronchoscopy (ENB) and radial probe endobronchial ultrasound (RP-EBUS) are essential bronchoscopic procedures for diagnosing peripheral lung lesions. Despite their individual advantages, the optimal circumstances for their combination remain uncertain. This single-center retrospective study enrolled 473 patients with 529 pulmonary nodules who underwent ENB and/or RP-EBUS biopsies between December 2021 and December 2022. Diagnostic yield was calculated using strict, intermediate, and liberal definitions. In the strict definition, only malignant and specific benign lesions were deemed diagnostic at the time of the index procedure. The intermediate and liberal definitions included additional results from the follow-up period. The diagnostic yield of the strict definition was not statistically different among the three groups (ENB/Combination/RP-EBUS 63.8%/64.2%/62.6%, p = 0.944). However, the diagnostic yield was superior in the ENB + RP-EBUS group for nodules with a bronchus type II or III and a solid part <20 mm (odds ratio 1.96, 95% confidence interval 1.09-3.53, p = 0.02). In terms of complications, bleeding was significantly higher in the ENB + RP-EBUS group (ENB/Combination/RP-EBUS 3.7% /6.2/0.6%, p = 0.002), but no major adverse event was observed. The combination of ENB and RP-EBUS enhanced the diagnostic yield for nodules with bronchus type II or III and solid part <20 mm, despite a slightly elevated risk of bleeding. Careful patient selection based on nodule characteristics is important to benefit from this combined approach.

Enhancing Early Lung Cancer Diagnosis: Predicting Lung Nodule Progression in Follow-Up Low-Dose CT Scan with Deep Generative Model.

Early diagnosis of lung cancer can significantly improve patient outcomes. We developed a Growth Predictive model based on the Wasserstein Generative Adversarial Network framework (GP-WGAN) to predict the nodule growth patterns in the follow-up LDCT scans. The GP-WGAN was trained with a training set (N = 776) containing 1121 pairs of nodule images with about 1-year intervals and deployed to an independent test set of 450 nodules on baseline LDCT scans to predict nodule images (GP-nodules) in their 1-year follow-up scans. The 450 GP-nodules were finally classified as malignant or benign by a lung cancer risk prediction (LCRP) model, achieving a test AUC of 0.827 ± 0.028, which was comparable to the AUC of 0.862 ± 0.028 achieved by the same LCRP model classifying real follow-up nodule images (p = 0.071). The net reclassification index yielded consistent outcomes (NRI = 0.04; p = 0.62). Other baseline methods, including Lung-RADS and the Brock model, achieved significantly lower performance (p < 0.05). The results demonstrated that the GP-nodules predicted by our GP-WGAN model achieved comparable performance with the nodules in the real follow-up scans for lung cancer diagnosis, indicating the potential to detect lung cancer earlier when coupled with accelerated clinical management versus the current approach of waiting until the next screening exam.

Nontraumatic intraoperative pulmonary nodule localization with laser guide stamping in a hybrid operating room.

Lung nodule localization using conventional image-guided video-assisted thoracoscopic surgery involves lung puncture, which increases the risk of needle-related complications. We aimed to evaluate the feasibility and safety of a single-stage non-invasive laser-guided stamping localization technique followed by resection under general anesthesia in a hybrid operating room. We retrospectively reviewed consecutive patients who underwent thoracoscopic surgery for small pulmonary nodules using laser-guided dye-stamping localization methods in a hybrid operating room between June 2023 and October 2023. During the study period, 18 patients with 20 lesions underwent single-stage intraoperative image-guided stamping video-assisted thoracoscopic surgery in the hybrid operating room. The median size of the nodules was 7.4 mm (interquartile range [IQR] 5.7-9.8 mm), and median distance from the pleural surface was 9.8 mm (IQR 7.7-14.6 mm). The median localization time was 26 min (IQR 23-34 min), whereas median operation time was 69 min (IQR 62-87 min). The total median operating room time was 146 min (IQR 136-157 min). Twelve patients underwent less than two cone-beam computed tomography scans, while 6 underwent more than two scans. The total median dose area product, including cone-beam computed tomography scans, was 5731.4 uGym2. No localization-related complications were observed, and the postoperative length of stay was 1 day (IQR 1-2 days). The single-stage image-guided pleural stamping technique for localizing small pulmonary nodules in a hybrid operating room is feasible and safe. Future research with larger cohorts is required to further explore the benefits of this workflow.

Recommendations to Improve Management of Incidental Pulmonary Nodules in Canada: Expert Panel Consensus.

Introduction: Incidental pulmonary nodules (IPN) are common radiologic findings, yet management of IPNs is inconsistent across Canada. This study aims to improve IPN management based on multidisciplinary expert consensus and provides recommendations to overcome patient and system-level barriers. Methods: A modified Delphi consensus technique was conducted. Multidisciplinary experts with extensive experience in lung nodule management in Canada were recruited to participate in the panel. A survey was administered in 3 rounds, using a 5-point Likert scale to determine the level of agreement (1 = extremely agree, 5 = extremely disagree). Results: Eleven experts agreed to participate in the panel; 10 completed all 3 rounds. Consensus was achieved for 183/217 (84.3%) statements. Panellists agreed that radiology reports should include a standardized summary of findings and follow-up recommendations for all nodule sizes (ie, <6, 6-8, and >8 mm). There was strong consensus regarding the importance of an automated system for patient follow-up and that leadership support for organizational change at the administrative level is of utmost importance in improving IPN management. There was no consensus on the need for standardized national referral pathways, development of new guidelines, or establishing a uniform picture archiving and communication system. Conclusion: Canadian IPN experts agree that improved IPN management should include standardized radiology reporting of IPNs, standardized and automated follow-up of patients with IPNs, guideline adherence and implementation, and leadership support for organizational change. Future research should focus on the implementation and long-term effectiveness of these recommendations in clinical practice.

CoF-DResNet: Cancer Metastasis Recognition Network based on Dynamic Coordinated Metabolic Attention and Structural Attention.

Cancer metastasis usually means that cancer cells spread to other tissues or organs, and the condition worsens. Identifying whether cancer has metastasized can help doctors infer the progression of a patient's condition and is an essential prerequisite for devising treatment plans. Fluorine 18 fluorodeoxyglucose positron emission tomography/computed tomography ( 18F -FDG PET/CT) is an advanced cancer diagnostic imaging technique that provides both metabolic and structural information. In cancer metastasis recognition tasks, effectively integrating metabolic and structural information stands as a key technology to enhance feature representation and recognition performance. This paper proposes a cancer metastasis identification network based on dynamic coordinated metabolic attention and structural attention to address these challenges. Specifically, metabolic and structural features are extracted by incorporating a dynamic coordinated attention module (DCAM) into two branches of ResNet networks, thereby amalgamating high metabolic spatial information from PET images with texture structure information from CT images, and dynamically adjusting this process through iterations. Next, to improve the efficacy of feature expression, a multi-receptive field feature fusion module (MRFM) is included in order to execute multi-receptive field fusion of semantic features. To validate the effectiveness of our proposed model, experiments were conducted on both a private lung lymph nodes dataset and a public soft tissue sarcomas dataset. The accuracy of our method reached 76.0% and 75.1% for the two datasets, respectively, demonstrating an improvement of 6.8% and 5.6% compared to ResNet, thus affirming the efficacy of our method.

Ultrathin Bronchoscopy Without Virtual Navigation for Diagnosis of Peripheral Lung Lesions

BackgroundThe increasing incidence of encountering lung nodules necessitates an ongoing search for improved diagnostic procedures. Various bronchoscopic technologies have been introduced or are in development, but further studies are needed to define a method that fits best in clinical practice and health care systems.Research questionHow do basic bronchoscopic tools including a combination of thin (outer diameter 4.2 mm) and ultrathin bronchoscopes (outer diameter 3.0 mm), radial endobronchial ultrasound (rEBUS) and fluoroscopy perform in peripheral pulmonary lesion diagnosis?Study Design and MethodsThis is a retrospective review of the performance of peripheral bronchoscopy using thin and ultrathin bronchoscopy with rEBUS and 2D fluoroscopy without a navigational system for evaluating peripheral lung lesions in a single academic medical center from 11/2015 to 1/2021. We used a strict definition for diagnostic yield and assessed the impact of different variables on diagnostic yield, specifically after employment of the ultrathin bronchoscope. Logistic regression models were employed to assess the independent associations of the most impactful variables.ResultsA total of 322 patients were included in this study. The median of the long axis diameter was 2.2 cm and the median distance of the center of the lesion from the visceral pleural surface was 1.9 cm. Overall diagnostic yield was 81.3% after employment of the ultrathin bronchoscope, with more detection of concentric rEBUS views (93% vs. 78%, p < 0.001). Sensitivity for detecting malignancy also increased from 60.5% to 74.7% (p = 0.033) after incorporating the ultrathin scope into practice, while bronchus sign and peripheral location of the lesion were not found to affect diagnostic yield. Concentric rEBUS view, solid appearance, upper/middle lobe location and larger size of the nodules were found to be independent predictors of successful achievement of diagnosis at bronchoscopy.Interpretation: This study demonstrates a high diagnostic yield of biopsy of lung lesions achieved by utilization of thin and ultrathin bronchoscopes. Direct visualization of small peripheral airways with simultaneous rEBUS confirmation increased localization rate of small lesions in a conventional bronchoscopy setting without virtual navigational planning.

The effect of noise removal filters on classifying different types of medical images

The number of studies on medical image processing with artificial intelligence-supported systems and the resulting diagnosis and classification of diseases has increased significantly as deep learning technology has developed. Doctors and specialists can save time and energy using the most successful deep learning models to classify medical images. This study aims to optimize various models and filters with grid search to increase classification success in medical images. Hybrid models were created by combining Vgg16, EfficientNetB0, ResNet50, and ResNet50v2 models with support vector machine classifiers. The models were trained and tested on three medical datasets: lung nodules, histopathological images for breast cancer as PatchCamelyon, and diagnosis of eye diseases as Retinal Fundus Multi-disease Image Dataset. The results demonstrated the optimized models' effectiveness compared to literature studies regarding various performance metrics. The proposed approaches present saliency maps and reveal the effects of filters on model success through comprehensive experiments on medical image datasets.

Pulmonary presentation and coil marcation of dirofilaria lung coin nodes - case report.

Clinicians face significant diagnostic challenges when dealing with nodal lesions of the lung. The majority of lung nodules are benign; however, asignificant number can be cancerous. Dirofilaria infestation should also be considered in the differential diagnosis of nodal lung lesions. Most patients with Dirofilaria infestation are asymptomatic. To determine the etiology of nodal lung infiltrations, procedures like radiological imaging, bronchoscopy, endobronchial ultrasound transbronchial needle aspiration (EBUS-TBNA), transthoracic biopsy (TTB), tumor markers, nonspecific serological tests, or surgical removal of the lung coin lesions can be performed. Coil marking of the smaller lung lesions under computed tomography (CT) guidance is ahelpful tool in determining the position and localization of the nodal lung lesions. Video-assisted thoracoscopic surgery (VATS) or robot-assisted thoracoscopic surgery (RATS) wedge resection is adiagnostic and therapeutic method of choice. No additional treatment of Dirofilaria other than wedge resection of the lung is necessary.

Radiomic score for lung nodules as a prognostic biomarker in locally advanced rectal cancer patients: A bi-institutional study.

Undetermined lung nodules are common in locally advanced rectal cancer (LARC) and lack precise risk stratification. This study aimed to develop a radiomic-based score (Rad-score) to distinguish metastasis and predict overall survival (OS) in patients with LARC and lung nodules. Retrospective data from two institutions (July 10, 2006-September 24, 2015) was used to develop and validate the Rad-score for distinguishing lung nodule malignancy. The prognostic value of the Rad-score was investigated in LARC cohorts, leading to the construction and validation of a clinical and radiomic score (Cli-Rad-score) that incorporates both clinical and radiomic information for the purpose of improving personalized clinical prognosis prediction. Descriptive statistics, survival analysis, and model comparison were performed to assess the results. The Rad-score demonstrated great performance in distinguishing malignancy, with C-index values of 0.793 [95% CI: 0.729-0.856] in the training set and 0.730 [95% CI: 0.666-0.874] in the validation set. In independent LARC cohorts, Rad-score validation achieved C-index values of 0.794 [95% CI: 0.737-0.851] and 0.747 [95% CI: 0.615-0.879]. Regarding prognostic prediction, Rad-score effectively stratified patients. Cli-Rad-score outperformed the clinicopathological information alone in risk stratification, as evidenced by significantly higher C-index values (0.735 vs. 0.695 in the internal set and 0.618 vs. 0.595 in the external set). CT-based radiomics could serve as a reliable and powerful tool for lung nodule malignancy distinction and prognostic prediction in LARC patients. Rad-score predicts prognosis independently. Incorporation of Cli-Rad-score significantly enhances the persionalized clinical prognostic capacity in LARC patients with lung nodules.

Adaptively learning visual model of surgical resection images via coarse-to-fine–grained strategy for pathological invasiveness prediction in indeterminate stage IA lung adenocarcinoma.

1563 Background: Indeterminate nodules with high risk of malignancy are suggested to receive video-assisted thoracoscopy biopsy. Frozen section assists surgeons to decide surgical strategies for stage IA lung adenocarcinoma (LUAD) intraoperatively, which is time-consuming and may misdiagnosis due to limited sampling and poor tissue quality. Methods: We prospectively enrolled stage IA LUAD patients underwent complete resection between June 2020 and September 2023 in Guangdong Provincial People's Hospital (GDPH), Affiliated Hospital of Guangdong Medical University, Meizhou People's Hospital. All the resected lung nodules were finally diagnosed as pre-invasive lesion (PIL), minimally invasive adenocarcinoma (MIA) or invasive lung adenocarcinoma (IAC) by FFPE pathological diagnosis. Images of the whole section of the lung nodules were taken by smartphones under natural lighting condition in operating theaters without shadowless surgical lights. Patients received preoperative anti-cancer therapy and with low-quality tumor section images were excluded. Images of nodules section, FS reports, FFPE diagnosis and clinical information were prospectively collected. Predictive artificial intelligence (AI) model was developed by a three-step process, which first predict the IAC and non-IAC region at coarse level, and identify fine level features through the high-risk region. The coarse and fine labels were then coordinated through risk ranking loss to reach a final predictive result. Results: We prospectively enrolled 1516 patients with 1638 indetermined lung nodules from preoperative chest CT following the inclusion and exclusion criteria, with 2438 images acquired from intraoperative nodule sections. The included patients have median age of 57 years old and 38.6% male patients. The pathological diagnosis for included lung nodules are 122 PIL nodules, 502 MIA nodules and 1014 stage IA LUAD. A multi-task artificial intelligence based on Coarse-to-Fine-Grained Strategy has been established to classify the indetermined lung nodules during the surgery. For binary classification, AI model reach an AUC of 0.86 in differentiation between IAC or non-IAC. For 3-classification level, AI model reach an AUC of 0.87 in differentiation between PIL, MIA or IAC. We further defined the IAC nodules into G1, G2 and G3 according to IASLC lung adenocarcinoma grading system. For 5-classification level, AI model reach an AUC of 0.86 in differentiation between PIL, MIA, IAC-G1, IAC-G2 and IAC-G3. Conclusions: Our AI models based on images of intraoperative surgical resection could effectively classify indetermined lung nodules, which can assist thoracic surgeons diagnose the nodules rapidly during the surging and decide the following surgical strategies. Clinical trial information: ChiCTR2300075999 .

Artificial intelligence-based prediction model of malignant lung nodules for preoperative planning.

8034 Background: The histopathological prediction of malignant lung nodules is crucial for preoperative planning, but it always remains not precise until the detailed pathological evaluation is performed after the surgery. Thus, to define the histology types (in situ adenocarcinoma (AIS), microinvasive adenocarcinoma (MIA), and invasive adenocarcinoma (IA)) of pulmonary adenocarcinoma appearing as lung nodules before the operation and reduce unnecessary invasive diagnosis and treatment operations, we developed a classification model of based on CT images. Methods: Patients who were diagnosed with pulmonary adenocarcinoma (tumor diameter ≤ 3cm) at Nanfang Hospital, Southern Medical University, China, were retrospectively enrolled, including detailed pathology results, computed tomography (CT) images, and basic information. Through training the model proposed in our previous study, we proposed a new model for defining histopathological types. Then, we used this data to train and compare the diagnostic efficacy of our model with another previously reported deep learning (DL) model (that was a segmentation - classification model with the 3D Unet++ - ResNet-50 combined model) and machine learning (ML) model (that was a traditional model Grey Level Co-occurrence Matrices (GLCM) as the feature extractor and the support vector machine (SVM) as the classifier) in the same method. Results: A total of 2061 patients (with 395 of AIS, 334 of MIA and 1332 of IA) were retrospectively enrolled from January 2019 to April 2022. The data from 2/3 of the patients were randomly selected for training and another 1/3 for verifying. Through training and verifying, our model shows much better diagnostic efficacy, with areas under receiver operating characteristic (ROC) curves with 95% confidence interval (CI) of 0.97(0.96, 0.98), comparing with those of the ML model of 0.77(0.73, 0.81) and the DL model of (0.90(0.88, 0.92). The heatmap visualization of the proposed model shows the classification ability with prediction probabilities of our model clearly and understandably. Conclusions: In this study, we developed a novel artificial intelligence multi-task learning model to precisely predict the pathology type of malignant lung nodules with satisfying diagnostic efficacy, which might provide a new and reliable way for preoperative planning of lung nodules. Further external verifying should be done in the future. In addition, prediction of the growth patterns of pulmonary adenocarcinoma could be more helpful in addition to predicting histopathological types.

Factors associated with delays in care of suspicious lung nodules at an academic medical center.

e13705 Background: Shorter time to lung cancer diagnosis and treatment are associated with improved outcomes. Here, we assessed factors contributing to delays in each phase of care to identify targets for improving the timeliness of suspicious lung nodule care in a diverse patient population. Methods: We conducted a retrospective study of patients evaluated for suspicious lung nodules at an academic medical center from 2020 to 2022. Patients with presumed or proven stage I/II lung cancer were included. Patients with stage III/IV and non-primary lung cancers were excluded. Multivariable Cox regressions were performed. Results: Of 157 patients, mean age was 70±8.6 years, 59% were female, and 53% were Black. Most lung nodules were biopsied bronchoscopically (61%) or surgically (26%). Most (83%) nodules were stage I/II non-small cell lung cancer, 10.2% were benign and 4.4% were other. Treatment was surgery in 48% and stereotactic body radiotherapy (SBRT) in 41%. Median intervals were 4 (IQR 0-18) days from nodule identification to referral, 7 (IQR 3-14) days from referral to consultation, 20 (IQR 12-34) days from consultation to biopsy, 4 (IQR 2-6) days from biopsy to diagnosis, and 28 (IQR 8-43) days from diagnosis to treatment initiation. Multivariable Cox regression showed that longer time from referral to diagnosis (TTD) was associated with Black race, widowed status, and referral &gt; 1 year from the start of the COVID-19 pandemic. Longer time from referral to treatment (TTT) was associated with female gender, widowed status, 4th Area Deprivation Index (ADI) quartile, frailty, BMI &gt; 18.5, ECOG performance status &lt; 2, bronchoscopic biopsy, and initial treatment with SBRT. Longer time from consultation to biopsy was associated with female gender, widowed status, frailty, no cardiovascular disease, and living &gt;12.5 miles from the hospital. Longer time from diagnosis to treatment was associated with Medicaid, referral &gt; 1 year from the start of the pandemic, cardiovascular disease, and SBRT. Conclusions: The longest delays in lung nodule care are associated with time from consultation to biopsy and from diagnosis to treatment. Efforts to address disparities and modifiable factors could improve the timeliness of care for patients with suspicious lung nodules. [Table: see text]

Use of cell-free tumor DNA in early detection of lung cancer.

e20088 Background: USPSTF recommends annual screening for lung cancer with low-dose computed tomography (LDCT) in adults 50-80 years with 20 pack-year smoking history or current/former smoking in the past 15 years. This prospective study evaluated whether we could detect lung cancer at the time of LDCT screening from cell-free tumor DNA using Guardant Health’s LUNAR-2 assay. Methods: Kaiser Permanente Colorado (KPCO) members with upcoming LDCT screenings were invited via email/phone and recruited into two groups, (1) suspicious lung nodules or (2) no lung nodules. We limited enrollment to patients whose LDCT results were classified under the Lung-RADS (version 1.0) categories of 1-3 (recommended follow-up at 12-month intervals) or category 4 (“suspicious”; recommended follow-up at 3-month intervals). We excluded members with lung nodules &gt; 30mm, those with a history of any hematologic malignancy, history of invasive malignancy within the past 5 years, currently pregnant, or known dementia or cognitive impairment. After informed consent, Guardant Health LUNAR test collection kits were mailed to the participant with sample collection instructions provided. Participants were asked to have their blood drawn on the day of their LDCT visit, or no more than 14 days of the scan. We also collected survey and EHR data on smoking history, LDCT, and other associated factors. The KPCO tumor registry confirmed all reported cancers. Results: The study initiated in September 2020 and ended in 08/2022 with 982 individuals consented and blood samples collected. Participants were on average 68 years old, 56% female, 58% former smokers, and 85% had no personal history of cancer. A total of 15 lung cancer cases and 22 other cancer cases were diagnosed. Data analysis is ongoing and expected to be completed early 2024. Conclusions: This study will demonstrate sensitivity and specificity of the LUNAR-2 assay to detect lung cancer relative to standard of care screening modalities in high-risk populations. We will also evaluate whether factors such as comorbidities, pack year history, or medications affect assay accuracy. The impact of this research can inform blood-based early detection of lung cancer.

Integrating Machine Learning in Clinical Practice for Characterizing the Malignancy of Solitary Pulmonary Nodules in PET/CT Screening.

The study investigates the efficiency of integrating Machine Learning (ML) in clinical practice for diagnosing solitary pulmonary nodules' (SPN) malignancy. Patient data had been recorded in the Department of Nuclear Medicine, University Hospital of Patras, in Greece. A dataset comprising 456 SPN characteristics extracted from CT scans, the SUVmax score from the PET examination, and the ultimate outcome (benign/malignant), determined by patient follow-up or biopsy, was used to build the ML classifier. Two medical experts provided their malignancy likelihood scores, taking into account the patient's clinical condition and without prior knowledge of the true label of the SPN. Incorporating human assessments into ML model training improved diagnostic efficiency by approximately 3%, highlighting the synergistic role of human judgment alongside ML. Under the latter setup, the ML model had an accuracy score of 95.39% (CI 95%: 95.29-95.49%). While ML exhibited swings in probability scores, human readers excelled in discerning ambiguous cases. ML outperformed the best human reader in challenging instances, particularly in SPNs with ambiguous probability grades, showcasing its utility in diagnostic grey zones. The best human reader reached an accuracy of 80% in the grey zone, whilst ML exhibited 89%. The findings underline the collaborative potential of ML and human expertise in enhancing SPN characterization accuracy and confidence, especially in cases where diagnostic certainty is elusive. This study contributes to understanding how integrating ML and human judgement can optimize SPN diagnostic outcomes, ultimately advancing clinical decision-making in PET/CT screenings.

Progress in Computer-aided Diagnosis of Lung Nodules based on CT Images

The objective of this study is to review the research progress of computer-aided diagnosis of pulmonary nodules based on CT images. This is done in order to address the challenges posed by the increasing incidence and difficulty of diagnosis of pulmonary nodules. Through an in-depth analysis of the key technologies, algorithms and application cases in the diagnosis of pulmonary nodules, we sought to identify how computer technology can be used to improve the accuracy and efficiency of diagnosis. The study found that CT images have the advantages of high resolution and multi-dimensional reconstruction in the detection of pulmonary nodules. However, interpreting CT images still requires specialized medical knowledge. Computer-aided diagnosis technology can assist doctors to identify lung nodules more accurately, especially for nodules with similar density to the surrounding tissue. This can improve the sensitivity and specificity of diagnosis. In conclusion, the computer-aided diagnosis system based on CT images provides substantial support for the accurate diagnosis of pulmonary nodules, which is beneficial for the improvement of patient health management and the formulation of treatment plans.

Automated 3D convolutional neural network architecture design using genetic algorithm for pulmonary nodule classification

Cancer of the lungs is considered one of the primary causes of death among patients globally. Early detection contributes significantly to the success of pulmonary cancer treatment. To aid the pulmonary nodule classification, many models for the analysis of medical image utilizing deep learning have been developed. Convolutional neural network (CNN) recently, has attained remarkable results in various image classification tasks. Nevertheless, the CNNs performance is heavily dependent on their architectures which still heavily reliant on human domain knowledge. This study introduces a cutting-edge approach that leverages genetic algorithms (GAs) to automatically design 3D CNN architectures for differentiation between benign and malignant pulmonary nodules. The suggested algorithm utilizes the dataset of lung nodule analysis 2016 (LUNA16) for evaluation. Notably, our approach achieved exceptional model accuracy, with evaluations on the testing dataset yielding up to 95.977%. Furthermore, the algorithm exhibited high sensitivity, showcasing its robust performance in distinguishing between benign and malignant nodules. Our findings demonstrate the outstanding capabilities of the proposed algorithm and show an outstanding performance and attain a state of art solution in lung nodule classification.

Development and validation of a preoperative CT‑based radiomics nomogram to differentiate tuberculosis granulomas from lung adenocarcinomas: an external validation study

BackgroundAn accurate and non-invasive approach is urgently needed to distinguish tuberculosis granulomas from lung adenocarcinomas. This study aimed to develop and validate a nomogram based on contrast enhanced-compute tomography (CE-CT) to preoperatively differentiate tuberculosis granuloma from lung adenocarcinoma appearing as solitary pulmonary solid nodules (SPSN).MethodsThis retrospective study analyzed 143 patients with lung adenocarcinoma (mean age: 62.4 ± 6.5 years; 54.5% female) and 137 patients with tuberculosis granulomas (mean age: 54.7 ± 8.2 years; 29.2% female) from two centers between March 2015 and June 2020. The training and internal validation cohorts included 161 and 69 patients (7:3 ratio) from center No.1, respectively. The external testing cohort included 50 patients from center No.2. Clinical factors and conventional radiological characteristics were analyzed to build independent predictors. Radiomics features were extracted from each CT-volume of interest (VOI). Feature selection was performed using univariate and multivariate logistic regression analysis, as well as the least absolute shrinkage and selection operator (LASSO) method. A clinical model was constructed with clinical factors and radiological findings. Individualized radiomics nomograms incorporating clinical data and radiomics signature were established to validate the clinical usefulness. The diagnostic performance was assessed using the receiver operating characteristic (ROC) curve analysis with the area under the receiver operating characteristic curve (AUC).ResultsOne clinical factor (CA125), one radiological characteristic (enhanced-CT value) and nine radiomics features were found to be independent predictors, which were used to establish the radiomics nomogram. The nomogram demonstrated better diagnostic efficacy than any single model, with respective AUC, accuracy, sensitivity, and specificity of 0.903, 0.857, 0.901, and 0.807 in the training cohort; 0.933, 0.884, 0.893, and 0.892 in the internal validation cohort; 0.914, 0.800, 0.937, and 0.735 in the external test cohort. The calibration curve showed a good agreement between prediction probability and actual clinical findings.ConclusionThe nomogram incorporating clinical factors, radiological characteristics and radiomics signature provides additional value in distinguishing tuberculosis granuloma from lung adenocarcinoma in patients with a SPSN, potentially serving as a robust diagnostic strategy in clinical practice.

VELOCITY-Lung substudy-03: A phase 2 study of neoadjuvant domvanalimab (dom)+zimberelimab (zim)+chemotherapy (chemo) or zim+chemo followed by adjuvant dom+zim or zim in patients with resectable stage II-III non-small cell lung cancer (NSCLC).

TPS8121 Background: Management of resectable NSCLC involves surgery with perioperative systemic therapy. Immune checkpoint inhibitors (CPIs) and targeted therapies are now used in the perioperative setting, leading to improved survival outcomes. The combination of dom (anti–T-cell immunoglobulin and ITIM domain [TIGIT]) and zim (anti–programmed death protein 1 [PD-1]) has shown promising antitumor activity with a manageable safety profile in metastatic NSCLC and may provide an opportunity to improve clinical outcomes in the perioperative setting in early-stage, resectable NSCLC. Substudy-03 of the VELOCITY-Lung platform study is evaluating novel perioperative treatment combinations in patients with newly diagnosed, resectable, stage II-III NSCLC. Methods: VELOCITY-Lung (NCT05633667) is an open-label, multicenter, phase 2 platform study with 3 ongoing substudies targeting the different NSCLC patient populations. Key eligibility criteria for Substudy-03 include age ≥ 18 years, histologically/cytologically confirmed stage II, IIIA, IIIB (T[3-4]N2) (per AJCC 8th ed.) squamous or nonsquamous NSCLC considered resectable with curative intent, no prior systemic or CPI therapy or radiotherapy, ECOG PS 0-1, any PD-1 ligand (PD-L1) status, and no actionable EGFR or ALK genomic alterations in patients with nonsquamous NSCLC. Randomization in Substudy-03 will be stratified by disease stage (II vs III) and baseline PD-L1 expression status (≥ 50% vs &lt; 50%). Current treatment groups in the neoadjuvant phase of the preliminary stage are: (A) dom+zim+platinum (Pt)-based chemo or (B) zim+Pt-based chemo for a maximum of 3 cycles; this will be followed by definitive surgery within 6 weeks of completing neoadjuvant treatment. Adjuvant treatment will include: (A) dom+zim or (B) zim and will begin within 12 weeks of surgery. Patients will continue to receive adjuvant treatment until disease recurrence, death, or unacceptable toxicity or to a maximum of 14 cycles. Dosing will be as follows: zim 360 mg intravenously (IV) once every 3 weeks (Q3W) and dom 1200 mg IV once Q3W (up to 3 cycles in the neoadjuvant phase and up to 14 cycles in the adjuvant phase). Pt-based chemo will be based on patient tumor histology (squamous vs nonsquamous). The primary endpoint is pathological complete response rate, defined as the percentage of patients with no residual invasive cancer in resected lung specimens and lymph nodes, assessed by local pathology review. Secondary endpoints include major pathological response rate (percentage of patients with ≤ 10% residual tumor in lung and lymph nodes at surgery), event-free survival, overall survival, and safety. Approximately 31 patients will be enrolled into each treatment group of the preliminary phase. The study is currently recruiting. Clinical trial information: NCT05633667 .