Risk factors of pneumothorax in computed tomography guided lung nodule marking using autologous blood: a retrospective study.

BACKGROUND: To investigate the risk factors of pneumothorax of using computed tomography (CT) guidance to inject autologous blood to locate isolated lung nodules. METHODS: In the First Hospital of Putian City, 92 cases of single small pulmonary nodules were retrospectively analyzed between November 2019 and March 2023. Before each surgery, autologous blood was injected, and the complications of each case, such as pneumothorax and pulmonary hemorrhage, were recorded. Patient sex, age, position at positioning, and nodule type, size, location, and distance from the visceral pleura were considered. Similarly, the thickness of the chest wall, the depth and duration of the needle-lung contact, the length of the positioning procedure, and complications connected to the patient's positioning were noted. Logistics single-factor and multi-factor variable analyses were used to identify the risk factors for pneumothorax. The multi-factor logistics analysis was incorporated into the final nomogram prediction model for modeling, and a nomogram was established. RESULTS: Logistics analysis suggested that the nodule size and the contact depth between the needle and lung tissue were independent risk factors for pneumothorax. CONCLUSION: The factors associated with pneumothorax after localization are smaller nodules and deeper contact between the needle and lung tissue.

Establishment and validation of multiclassification prediction models for pulmonary nodules based on machine learning.

BACKGROUND: Lung cancer is the leading cause of cancer-related death worldwide. This study aimed to establish novel multiclassification prediction models based on machine learning (ML) to predict the probability of malignancy in pulmonary nodules (PNs) and to compare with three published models. METHODS: Nine hundred fourteen patients with PNs were collected from four medical institutions (A, B, C and D), which were organized into tables containing clinical features, radiologic features and laboratory test features. Patients were divided into benign lesion (BL), precursor lesion (PL) and malignant lesion (ML) groups according to pathological diagnosis. Approximately 80% of patients in A (total/male: 632/269, age: 57.73 ± 11.06) were randomly selected as a training set; the remaining 20% were used as an internal test set; and the patients in B (total/male: 94/53, age: 60.04 ± 11.22), C (total/male: 94/47, age: 59.30 ± 9.86) and D (total/male: 94/61, age: 62.0 ± 11.09) were used as an external validation set. Logical regression (LR), decision tree (DT), random forest (RF) and support vector machine (SVM) were used to establish prediction models. Finally, the Mayo model, Peking University People's Hospital (PKUPH) model and Brock model were externally validated in our patients. RESULTS: The AUC values of RF model for MLs, PLs and BLs were 0.80 (95% CI: 0.73-0.88), 0.90 (95% CI: 0.82-0.99) and 0.75 (95% CI: 0.67-0.88), respectively. The weighted average AUC value of the RF model for the external validation set was 0.71 (95% CI: 0.67-0.73), and its AUC values for MLs, PLs and BLs were 0.71 (95% CI: 0.68-0.79), 0.98 (95% CI: 0.88-1.07) and 0.68 (95% CI: 0.61-0.74), respectively. The AUC values of the Mayo model, PKUPH model and Brock model were 0.68 (95% CI: 0.62-0.74), 0.64 (95% CI: 0.58-0.70) and 0.57 (95% CI: 0.49-0.65), respectively. CONCLUSIONS: The RF model performed best, and its predictive performance was better than that of the three published models, which may provide a new noninvasive method for the risk assessment of PNs.

[Clinical Application of Robotic Assisted Bronchoscopy 
in Peripheral Pulmonary Nodule Biopsy].

With the popularization of chest computed tomography (CT) lung cancer screening, the detection rate of peripheral pulmonary nodules is increasing day by day. Some patients could make clear diagnoses and receive early treatment by obtaining biopsy specimens. Transbronchial lung biopsy (TBLB) is one of the non-surgical biopsy methods for peripheral pulmonary nodules, which has less trauma and lower incidence of complications compared to percutaneous thoracic needle biopsy (PTNB). However, the diagnostic rate of TBLB is about 70%, which is still inferior to that of PTNB, which is about 90%. Since 2018, robot assisted bronchoscopy systems have been applied in clinical practice. This article reviews their application in further improving the diagnostic rate of peripheral pulmonary nodules by TBLB.
.

Value of turbo spin echo-based diffusion-weighted imaging in the differential diagnosis of benign and malignant solitary pulmonary lesions.

To quantitatively assess the diagnostic efficacy of multiple parameters derived from multi-b-value diffusion-weighted imaging (DWI) using turbo spin echo (TSE)-based acquisition techniques in patients with solitary pulmonary lesions (SPLs). A total of 105 patients with SPLs underwent lung DWI using single-shot TSE-based acquisition techniques and multiple b values. The apparent diffusion coefficient (ADC), intravoxel incoherent motion (IVIM) parameters, and lesion-to-spinal cord signal intensity ratio (LSR), were analyzed to compare the benign and malignant groups using the Mann-Whitney U test and receiver operating characteristic analysis. The Dstar values observed in lung cancer were slightly lower than those observed in pulmonary benign lesions (28.164 ± 31.950 versus 32.917 ± 34.184; Z = -2.239, p = 0.025). The LSR values were significantly higher in lung cancer than in benign lesions (1.137 ± 0.581 versus 0.614 ± 0.442; Z = - 4.522, p < 0.001). Additionally, the ADC800, ADCtotal, and D values were all significantly lower in lung cancer than in the benign lesions (Z = - 5.054, -5.370, and -6.047, respectively, all p < 0.001), whereas the f values did not exhibit any statistically significant difference between the two groups. D had the highest area under the curve (AUC = 0.887), followed by ADCtotal (AUC = 0.844), ADC800 (AUC = 0.824), and LSR (AUC = 0.789). The LSR, ADC800, ADCtotal, and D values did not differ statistically significantly in diagnostic effectiveness. Lung DWI using TSE is feasible for differentiating SPLs. The LSR method, conventional DWI, and IVIM have comparable diagnostic efficacy for assessing SPLs.

Attention pyramid pooling network for artificial diagnosis on pulmonary nodules.

The development of automated tools using advanced technologies like deep learning holds great promise for improving the accuracy of lung nodule classification in computed tomography (CT) imaging, ultimately reducing lung cancer mortality rates. However, lung nodules can be difficult to detect and classify, from CT images since different imaging modalities may provide varying levels of detail and clarity. Besides, the existing convolutional neural network may struggle to detect nodules that are small or located in difficult-to-detect regions of the lung. Therefore, the attention pyramid pooling network (APPN) is proposed to identify and classify lung nodules. First, a strong feature extractor, named vgg16, is used to obtain features from CT images. Then, the attention primary pyramid module is proposed by combining the attention mechanism and pyramid pooling module, which allows for the fusion of features at different scales and focuses on the most important features for nodule classification. Finally, we use the gated spatial memory technique to decode the general features, which is able to extract more accurate features for classifying lung nodules. The experimental results on the LIDC-IDRI dataset show that the APPN can achieve highly accurate and effective for classifying lung nodules, with sensitivity of 87.59%, specificity of 90.46%, accuracy of 88.47%, positive predictive value of 95.41%, negative predictive value of 76.29% and area under receiver operating characteristic curve of 0.914.

Impact of different respiratory gating methods on target delineation and a radiotherapy plan for solitary pulmonary tumors.

BACKGROUND AND PURPOSE: Respiratory movement has an important impact on the radiotherapy for lung tumor. Respiratory gating technology is helpful to improve the accuracy of target delineation. This study investigated the value of prospective and retrospective respiratory gating simulations in target delineation and radiotherapy plan design for solitary pulmonary tumors (SPTs) in radiotherapy. METHODS: The enrolled patients underwent CT simulation with three-dimensional (3D) CT non gating, prospective respiratory gating, and retrospective respiratory gating simulation. The target volumes were delineated on three sets of CT images, and radiotherapy plans were prepared accordingly. Tumor displacements and movement information obtained using the two respiratory gating approaches, as well as the target volumes and dosimetry parameters in the radiotherapy plan were compared. RESULTS: No significant difference was observed in tumor displacement measured using the two gating methods (p > 0.05). However, the internal gross tumor volumes (IGTVs), internal target volumes (ITVs), and planning target volumes (PTVs) based on the retrospective respiratory gating simulation were larger than those obtained using prospective gating (group A: pIGTV = 0.041, pITV = 0.003, pPTV = 0.008; group B: pIGTV = 0.025, pITV = 0.039, pPTV = 0.004). The two-gating PTVs were both smaller than those delineated on 3D non gating images (p < 0.001). V5Gy, V10Gy, V20Gy, V30Gy, and mean lung dose in the two gated radiotherapy plans were lower than those in the 3D non gating plan (p < 0.001); however, no significant difference was observed between the two gating plans (p > 0.05). CONCLUSIONS: The application of respiratory gating could reduce the target volume and the radiation dose that the normal lung tissue received. Compared to prospective respiratory gating, the retrospective gating provides more information about tumor movement in PTV.

Imaging of Solid Pulmonary Nodules.

Early detection with accurate classification of solid pulmonary nodules is critical in reducing lung cancer morbidity and mortality. Computed tomography (CT) remains the most widely used imaging examination for pulmonary nodule evaluation; however, other imaging modalities, such as PET/CT and MRI, are increasingly used for nodule characterization. Current advances in solid nodule imaging are largely due to developments in machine learning, including automated nodule segmentation and computer-aided detection. This review explores current multi-modality solid pulmonary nodule detection and characterization with discussion of radiomics and risk prediction models.

Subsolid Nodules: Significance and Current Understanding.

Subsolid nodules are heterogeneously appearing and behaving entities, commonly encountered incidentally and in high-risk populations. Accurate characterization of subsolid nodules, and application of evolving surveillance guidelines, facilitates evidence-based and multidisciplinary patient-centered management.

Impact of deep learning image reconstruction on volumetric accuracy and image quality of pulmonary nodules with different morphologies in low-dose CT.

BACKGROUND: This study systematically compares the impact of innovative deep learning image reconstruction (DLIR, TrueFidelity) to conventionally used iterative reconstruction (IR) on nodule volumetry and subjective image quality (IQ) at highly reduced radiation doses. This is essential in the context of low-dose CT lung cancer screening where accurate volumetry and characterization of pulmonary nodules in repeated CT scanning are indispensable. MATERIALS AND METHODS: A standardized CT dataset was established using an anthropomorphic chest phantom (Lungman, Kyoto Kaguku Inc., Kyoto, Japan) containing a set of 3D-printed lung nodules including six diameters (4 to 9 mm) and three morphology classes (lobular, spiculated, smooth), with an established ground truth. Images were acquired at varying radiation doses (6.04, 3.03, 1.54, 0.77, 0.41 and 0.20 mGy) and reconstructed with combinations of reconstruction kernels (soft and hard kernel) and reconstruction algorithms (ASIR-V and DLIR at low, medium and high strength). Semi-automatic volumetry measurements and subjective image quality scores recorded by five radiologists were analyzed with multiple linear regression and mixed-effect ordinal logistic regression models. RESULTS: Volumetric errors of nodules imaged with DLIR are up to 50% lower compared to ASIR-V, especially at radiation doses below 1 mGy and when reconstructed with a hard kernel. Also, across all nodule diameters and morphologies, volumetric errors are commonly lower with DLIR. Furthermore, DLIR renders higher subjective IQ, especially at the sub-mGy doses. Radiologists were up to nine times more likely to score the highest IQ-score to these images compared to those reconstructed with ASIR-V. Lung nodules with irregular margins and small diameters also had an increased likelihood (up to five times more likely) to be ascribed the best IQ scores when reconstructed with DLIR. CONCLUSION: We observed that DLIR performs as good as or even outperforms conventionally used reconstruction algorithms in terms of volumetric accuracy and subjective IQ of nodules in an anthropomorphic chest phantom. As such, DLIR potentially allows to lower the radiation dose to participants of lung cancer screening without compromising accurate measurement and characterization of lung nodules.

An 18F-FDG-PET/CT-based radiomics signature for estimating malignance probability of solitary pulmonary nodule.

BACKGROUND: Some solitary pulmonary nodules (SPNs) as early manifestations of lung cancer, it is difficult to determine its nature, which brings great trouble to clinical diagnosis and treatment. Radiomics can deeply explore the essence of images and provide clinical decision support for clinicians. The purpose of our study was to explore the effect of positron emission tomography (PET) with 2-deoxy-2-[fluorine-18] fluoro-d-glucose integrated with computed tomography (CT; 18F-FDG-PET/CT) combined with radiomics for predicting probability of malignancy of SPNs. METHODS: We retrospectively enrolled 190 patients with SPNs confirmed by pathology from January 2013 to December 2019 in our hospital. SPNs were benign in 69 patients and malignant in 121 patients. Patients were randomly divided into a training or testing group at a ratio of 7:3. Three-dimensional regions of interest (ROIs) were manually outlined on PET and CT images, and radiomics features were extracted. Synthetic minority oversampling technique (SMOTE) method was used to balance benign and malignant samples to a ratio of 1:1. In the training group, least absolute shrinkage and selection operator (LASSO) regression analyses and Spearman correlation analyses were used to select the strongest radiomics features. Three models including PET model, CT model, and joint model were constructed using multivariate logistic regression analysis. Receiver operating characteristic (ROC) curves, calibration curves, and decision curves were plotted to evaluate diagnostic efficiency, calibration degree, and clinical usefulness of all models in training and testing groups. RESULTS: The estimative effectiveness of the joint model was superior to the CT or PET model alone in the training and testing groups. For the joint model, CT model, and PET model, area under the ROC curve was 0.929, 0.819, 0.833 in the training group, and 0.844, 0.759, 0.748 in the testing group, respectively. Calibration and decision curves showed good fit and clinical usefulness for the joint model in both training and testing groups. CONCLUSION: Radiomics models constructed by combining PET and CT radiomics features are valuable for distinguishing benign and malignant SPNs. The combined effect is superior to qualitative diagnoses with CT or PET radiomics models alone.

LNDb v4: pulmonary nodule annotation from medical reports.

Given the high prevalence of lung cancer, an accurate diagnosis is crucial. In the diagnosis process, radiologists play an important role by examining numerous radiology exams to identify different types of nodules. To aid the clinicians' analytical efforts, computer-aided diagnosis can streamline the process of identifying pulmonary nodules. For this purpose, medical reports can serve as valuable sources for automatically retrieving image annotations. Our study focused on converting medical reports into nodule annotations, matching textual information with manually annotated data from the Lung Nodule Database (LNDb)-a comprehensive repository of lung scans and nodule annotations. As a result of this study, we have released a tabular data file containing information from 292 medical reports in the LNDb, along with files detailing nodule characteristics and corresponding matches to the manually annotated data. The objective is to enable further research studies in lung cancer by bridging the gap between existing reports and additional manual annotations that may be collected, thereby fostering discussions about the advantages and disadvantages between these two data types.

Diagnostic yield using electromagnetic navigation bronchoscopy for peripheral pulmonary nodules <2 cm.

BACKGROUND: Although electromagnetic navigation bronchoscopy (ENB) is highly sensitive in the diagnosis of peripheral pulmonary nodules (PPNs), its diagnostic yield for subgroups of smaller PPNs is under evaluation. OBJECTIVES: Diagnostic yield evaluation of biopsy using ENB for PPNs <2 cm. DESIGN: The diagnostic yield, sensitivity, specificity, positive predictive value, and negative predictive value of the ENB-mediated biopsy for PPNs were evaluated. METHODS: Patients who had PPNs with diameters <2 cm and underwent ENB-mediated biopsy between May 2015 and February 2020 were consecutively enrolled. The final diagnosis was made via pathological examination after surgery. RESULTS: A total of 82 lesions from 65 patients were analyzed. The median tumor size was 11 mm. All lesions were subjected to ENB-mediated biopsy, of which 29 and 53 were classified as malignant and benign, respectively. Subsequent segmentectomy, lobectomy, or wedge resection, following pathological examinations were performed on 64 nodules from 57 patients. The overall sensitivity, specificity, positive predictive value, and negative predictive value for nodules <2 cm were 53.3%, 91.7%, 92.3%, and 51.2%, respectively. The receiver operating curve showed an area under the curve of 0.721 (p < 0.001). Additionally, the sensitivity, specificity, positive predictive value, and negative predictive value were 62.5%, 100%, 100%, and 42.9%, respectively, for nodules with diameters equal to or larger than 1 cm; and 30.8%, 86.7%, 66.7%, and 59.1%, respectively, for nodules less than 1 cm. In the subgroup analysis, neither the lobar location nor the distance of the PPNs to the pleura affected the accuracy of the ENB diagnosis. However, the spiculated sign had a negative impact on the accuracy of the ENB biopsy (p = 0.010). CONCLUSION: ENB has good specificity and positive predictive value for diagnosing PPNs <2 cm; however, the spiculated sign may negatively affect ENB diagnostic accuracy. In addition, the diagnostic reliability may only be limited to PPNs equal to or larger than 1 cm.

Effect of emphysema on AI software and human reader performance in lung nodule detection from low-dose chest CT.

BACKGROUND: Emphysema influences the appearance of lung tissue in computed tomography (CT). We evaluated whether this affects lung nodule detection by artificial intelligence (AI) and human readers (HR). METHODS: Individuals were selected from the "Lifelines" cohort who had undergone low-dose chest CT. Nodules in individuals without emphysema were matched to similar-sized nodules in individuals with at least moderate emphysema. AI results for nodular findings of 30-100 mm3 and 101-300 mm3 were compared to those of HR; two expert radiologists blindly reviewed discrepancies. Sensitivity and false positives (FPs)/scan were compared for emphysema and non-emphysema groups. RESULTS: Thirty-nine participants with and 82 without emphysema were included (n = 121, aged 61 ± 8 years (mean ± standard deviation), 58/121 males (47.9%)). AI and HR detected 196 and 206 nodular findings, respectively, yielding 109 concordant nodules and 184 discrepancies, including 118 true nodules. For AI, sensitivity was 0.68 (95% confidence interval 0.57-0.77) in emphysema versus 0.71 (0.62-0.78) in non-emphysema, with FPs/scan 0.51 and 0.22, respectively (p = 0.028). For HR, sensitivity was 0.76 (0.65-0.84) and 0.80 (0.72-0.86), with FPs/scan of 0.15 and 0.27 (p = 0.230). Overall sensitivity was slightly higher for HR than for AI, but this difference disappeared after the exclusion of benign lymph nodes. FPs/scan were higher for AI in emphysema than in non-emphysema (p = 0.028), while FPs/scan for HR were higher than AI for 30-100 mm3 nodules in non-emphysema (p = 0.009). CONCLUSIONS: AI resulted in more FPs/scan in emphysema compared to non-emphysema, a difference not observed for HR. RELEVANCE STATEMENT: In the creation of a benchmark dataset to validate AI software for lung nodule detection, the inclusion of emphysema cases is important due to the additional number of FPs. KEY POINTS: • The sensitivity of nodule detection by AI was similar in emphysema and non-emphysema. • AI had more FPs/scan in emphysema compared to non-emphysema. • Sensitivity and FPs/scan by the human reader were comparable for emphysema and non-emphysema. • Emphysema and non-emphysema representation in benchmark dataset is important for validating AI.

Comparison between one-stage and two-stage CT-guided localization of lung nodules with surgical resection: a single medical center experience.

BACKGROUND: This retrospective study aimed to compare the efficacy and safety of one-stage computed tomography (OSCT)- to that of two-stage computed tomography (TSCT)-guided localization for the surgical removal of small lung nodules. METHODS: We collected data from patients with ipsilateral pulmonary nodules who underwent localization before surgical removal at Veteran General Hospital Kaohsiung between October 2017 and January 2022. The patients were divided into the OSCT and TSCT groups. RESULTS: We found that OSCT significantly reduced the localization time and risky time compared to TSCT, and the success rate of localization and incidence of pneumothorax were similar in both groups. However, the time spent under general anesthesia was longer in the OSCT group than in the TSCT group. CONCLUSIONS: The OSCT-guided approach to localize pulmonary nodules in hybrid operation room is a safe and effective technique for the surgical removal of small lung nodules.

The use of diagnostic complex robotic-assisted segmentectomy in the management of incidental and screen-detected pulmonary nodules.

OBJECTIVES: Robotic-assisted thoracoscopic surgery (RATS) facilitates complex pulmonary segmentectomy which offers one-stage diagnostic and therapeutic management of small pulmonary nodules. We aimed to explore the potential advantages of a faster, simplified pathway and earlier diagnosis against the disadvantages of unnecessary morbidity in benign cases. METHODS: In an observational study, patients with small, solitary pulmonary nodules deemed suspicious of malignancy by a multidisciplinary team were offered surgery without a pre or intraoperative biopsy. We report our initial experience with RATS complex segmentectomy (using >1 parenchymal staple line) to preserve as much functioning lung tissue as possible. RESULTS: Over a 4-year period, 245 RATS complex segmentectomies were performed; 140 right: 105 left. A median of 2 (1-4) segments was removed. There was no in-hospital mortality and no requirement for postoperative ventilation. Complications were reported in 63 (25.7%) cases, of which 36 (57.1%) were hospital-acquired pneumonia. A malignant diagnosis was found in 198 (81%) patients and a benign diagnosis in 47 (19%). The malignant diagnoses included: adenocarcinoma in 136, squamous carcinoma in 31 and carcinoid tumour in 15. The most frequent benign diagnosis was granulomatous inflammation in 18 cases. CONCLUSIONS: RATS complex segmentectomy offers a precise, safe and effective one-stop therapeutic biopsy in incidental and screen-detected pulmonary nodules.

Application of 18F-FDG PET/CT imaging radiomics in the differential diagnosis of single-nodule pulmonary metastases and second primary lung cancer in patients with colorectal cancer.

OBJECTIVE: It is crucially essential to differentially diagnose single-nodule pulmonary metastases (SNPMs) and second primary lung cancer (SPLC) in patients with colorectal cancer (CRC), which has important clinical implications for treatment strategies. In this study, we aimed to establish a feasible differential diagnosis model by combining 18F-fluorodeoxyglucose positron-emission tomography (18F-FDG PET) radiomics, computed tomography (CT) radiomics, and clinical features. MATERIALS AND METHODS: CRC patients with SNPM or SPLC who underwent 18F-FDG PET/CT from January 2013 to July 2022 were enrolled in this retrospective study. The radiomic features were extracted by manually outlining the lesions on PET/CT images, and the radiomic modeling was realized by various screening methods and classifiers. In addition, clinical features were analyzed by univariate analysis and logistic regression (LR) analysis to be included in the combined model. Finally, the diagnostic performances of these models were illustrated by the receiver operating characteristic (ROC) curves and the area under the curve (AUC). RESULTS: We studied data from 61 patients, including 36 SNPMs and 25 SPLCs, with an average age of 65.56 ± 10.355 years. Spicule sign and ground-glass opacity (GGO) were significant independent predictors of clinical features (P = 0.012 and P < 0.001, respectively) to build the clinical model. We achieved a PET radiomic model (AUC = 0.789), a CT radiomic model (AUC = 0.818), and a PET/CT radiomic model (AUC = 0.900). The PET/CT radiomic models were combined with the clinical model, and a well-performing model was established by LR analysis (AUC = 0.940). CONCLUSIONS: For CRC patients, the radiomic models we developed had good performance for the differential diagnosis of SNPM and SPLC. The combination of radiomic and clinical features had better diagnostic value than a single model.

[The role of endoscopy in the management of peripheral pulmonary nodules, part 2: Treatment]. / Place de l'endoscopie dans la gestion des nodules pulmonaires périphériques, partie 2 : traitement.

The management of peripheral lung nodules is challenging, requiring specialized skills and sophisticated technologies. The diagnosis now appears accessible to advanced endoscopy (see Part 1), which can also guide treatment of these nodules; this second part provides an overview of endoscopy techniques that can enhance surgical treatment through preoperative marking, and stereotactic radiotherapy treatment through fiduciary marker placement. Finally, we will discuss how, in the near future, these advanced endoscopic techniques will help to implement ablation strategy.

Robust deep learning from incomplete annotation for accurate lung nodule detection.

Deep learning plays a significant role in the detection of pulmonary nodules in low-dose computed tomography (LDCT) scans, contributing to the diagnosis and treatment of lung cancer. Nevertheless, its effectiveness often relies on the availability of extensive, meticulously annotated dataset. In this paper, we explore the utilization of an incompletely annotated dataset for pulmonary nodules detection and introduce the FULFIL (Forecasting Uncompleted Labels For Inexpensive Lung nodule detection) algorithm as an innovative approach. By instructing annotators to label only the nodules they are most confident about, without requiring complete coverage, we can substantially reduce annotation costs. Nevertheless, this approach results in an incompletely annotated dataset, which presents challenges when training deep learning models. Within the FULFIL algorithm, we employ Graph Convolution Network (GCN) to discover the relationships between annotated and unannotated nodules for self-adaptively completing the annotation. Meanwhile, a teacher-student framework is employed for self-adaptive learning using the completed annotation dataset. Furthermore, we have designed a Dual-Views loss to leverage different data perspectives, aiding the model in acquiring robust features and enhancing generalization. We carried out experiments using the LUng Nodule Analysis (LUNA) dataset, achieving a sensitivity of 0.574 at a False positives per scan (FPs/scan) of 0.125 with only 10% instance-level annotations for nodules. This performance outperformed comparative methods by 7.00%. Experimental comparisons were conducted to evaluate the performance of our model and human experts on test dataset. The results demonstrate that our model can achieve a comparable level of performance to that of human experts. The comprehensive experimental results demonstrate that FULFIL can effectively leverage an incomplete pulmonary nodule dataset to develop a robust deep learning model, making it a promising tool for assisting in lung nodule detection.