Predicting the invasiveness of lung adenocarcinomas appearing as ground-glass nodule on CT scan using multi-task learning and deep radiomics

CT-based radiomics combined with clinical features for invasiveness prediction and pathological subtypes classification of subsolid pulmonary nodules

Purpose

To construct optimal models for predicting the invasiveness and pathological subtypes of subsolid nodules (SSNs) based on CT radiomics and clinical features.

Materials and Methods

This study was a retrospective study involving two centers. A total of 316 patients with 353 SSNs confirmed as atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC) were included from January 2019 to February 2023. Models based on CT radiomics and clinical features were constructed for classification of AAH/AIS and MIA, MIA and IAC, as well as lepidic-predominant adenocarcinoma (LPA) and acinar-predominant adenocarcinoma (APA). Receiver operating characteristic (ROC) curve was used to evaluate the model performance. Finally, the nomograms based on the optimal models were established.

Results

The nomogram based on the combined model (AAH/AIS versus MIA) consisting of lobulation, the GGN-vessel relationship, diameter, CT value, consolidation tumor ratio (CTR) and rad-score performed the best (AUC=0.841), while age, CT value, CTR and rad-score were the significant features for distinguishing MIA from IAC, the nomogram based on these features performed the best (AUC=0.878). There were no significant differences in clinical features between LPA and APA, while the radiomics model based on rad-score showed good performance for distinguishing LPA from APA (AUC=0.926).

Conclusions

The nomograms based on radiomics and clinical features could predict the invasiveness of SSNs accurately. Moreover, radiomics models showed good performance in distinguishing LPA from APA.

Res-TransNet: A Hybrid deep Learning Network for Predicting Pathological Subtypes of lung Adenocarcinoma in CT Images

This study aims to develop a CT-based hybrid deep learning network to predict pathological subtypes of early-stage lung adenocarcinoma by integrating residual network (ResNet) with Vision Transformer (ViT). A total of 1411 pathologically confirmed ground-glass nodules (GGNs) retrospectively collected from two centers were used as internal and external validation sets for model development. 3D ResNet and ViT were applied to investigate two deep learning frameworks to classify three subtypes of lung adenocarcinoma namely invasive adenocarcinoma (IAC), minimally invasive adenocarcinoma and adenocarcinoma in situ, respectively. To further improve the model performance, four Res-TransNet based models were proposed by integrating ResNet and ViT with different ensemble learning strategies. Two classification tasks involving predicting IAC from Non-IAC (Task1) and classifying three subtypes (Task2) were designed and conducted in this study. For Task 1, the optimal Res-TransNet model yielded area under the receiver operating characteristic curve (AUC) values of 0.986 and 0.933 on internal and external validation sets, which were significantly higher than that of ResNet and ViT models (p < 0.05). For Task 2, the optimal fusion model generated the accuracy and weighted F1 score of 68.3% and 66.1% on the external validation set. The experimental results demonstrate that Res-TransNet can significantly increase the classification performance compared with the two basic models and have the potential to assist radiologists in precision diagnosis.

Integrated whole-exome and bulk transcriptome sequencing delineates the dynamic evolution from preneoplasia to invasive lung adenocarcinoma featured with ground-glass nodules

OBJECTIVE

The genomic and molecular ecology involved in the stepwise continuum progression of lung adenocarcinoma (LUAD) from adenocarcinoma in situ (AIS) to minimally invasive adenocarcinoma (MIA) and subsequent invasive adenocarcinoma (IAC) remains unclear and requires further elucidation. We aimed to characterize gene mutations and expression landscapes, and explore the association between differentially expressed genes (DEGs) and significantly mutated genes (SMGs) during the dynamic evolution from AIS to IAC.

METHODS

Thirty-five patients with ground-glass nodules (GGNs) lung adenocarcinomas were enrolled. Whole-exome sequencing (WES) and transcriptome sequencing (RNA-Seq) were conducted on all patients, encompassing both tumor samples and corresponding noncancerous tissues. Data obtained from WES and RNA-Seq were subsequently analyzed.

RESULTS

The findings from WES delineated that the predominant mutations were observed in EGFR (49%) and ANKRD36C (17%). SMGs, including EGFR and RBM10, were associated with the dynamic evolution from AIS to IAC. Meanwhile, DEGs, including GPR143, CCR9, ADAMTS16, and others were associated with the entire process of invasive LUAD. We found that the signaling pathways related to cell migration and invasion were upregulated, and the signaling pathways of angiogenesis were downregulated across the pathological stages. Furthermore, we found that the messenger RNA (mRNA) levels of FAM83A, MAL2, DEPTOR, and others were significantly correlated with CNVs. Gene set enrichment analysis (GSEA) showed that heme metabolism and cholesterol homeostasis pathways were significantly upregulated in patients with EGFR/RBM10 co-mutations, and these patients may have poorer overall survival than those with EGFR mutations. Based on the six calculation methods for the immune infiltration score, NK/CD8+ T cells decreased, and Treg/B cells increased with the progression of early LUAD.

CONCLUSIONS

Our findings offer valuable insights into the unique genomic and molecular features of LUAD, facilitating the identification and advancement of precision medicine strategies targeting the invasive progression of LUAD from AIS to IAC.

Development of a modified 3D region proposal network for lung nodule detection in computed tomography scans: a secondary analysis of lung nodule datasets

BACKGROUND

Low-dose computed tomography (LDCT) has been shown useful in early lung cancer detection. This study aimed to develop a novel deep learning model for detecting pulmonary nodules on chest LDCT images.

METHODS

In this secondary analysis, three lung nodule datasets, including Lung Nodule Analysis 2016 (LUNA16), Lung Nodule Received Operation (LNOP), and Lung Nodule in Health Examination (LNHE), were used to train and test deep learning models. The 3D region proposal network (RPN) was modified via a series of pruning experiments for better predictive performance. The performance of each modified deep leaning model was evaluated based on sensitivity and competition performance metric (CPM). Furthermore, the performance of the modified 3D RPN trained on three datasets was evaluated by 10-fold cross validation. Temporal validation was conducted to assess the reliability of the modified 3D RPN for detecting lung nodules.

RESULTS

The results of pruning experiments indicated that the modified 3D RPN composed of the Cross Stage Partial Network (CSPNet) approach to Residual Network (ResNet) Xt (CSP-ResNeXt) module, feature pyramid network (FPN), nearest anchor method, and post-processing masking, had the optimal predictive performance with a CPM of 92.2%. The modified 3D RPN trained on the LUNA16 dataset had the highest CPM (90.1%), followed by the LNOP dataset (CPM: 74.1%) and the LNHE dataset (CPM: 70.2%). When the modified 3D RPN trained and tested on the same datasets, the sensitivities were 94.6%, 84.8%, and 79.7% for LUNA16, LNOP, and LNHE, respectively. The temporal validation analysis revealed that the modified 3D RPN tested on LNOP test set achieved a CPM of 71.6% and a sensitivity of 85.7%, and the modified 3D RPN tested on LNHE test set had a CPM of 71.7% and a sensitivity of 83.5%.

CONCLUSION

A modified 3D RPN for detecting lung nodules on LDCT scans was designed and validated, which may serve as a computer-aided diagnosis system to facilitate lung nodule detection and lung cancer diagnosis.

A novel image deep learning-based sub-centimeter pulmonary nodule management algorithm to expedite resection of the malignant and avoid over-diagnosis of the benign

OBJECTIVES

With the popularization of chest computed tomography (CT) screening, there are more sub-centimeter (≤ 1 cm) pulmonary nodules (SCPNs) requiring further diagnostic workup. This area represents an important opportunity to optimize the SCPN management algorithm avoiding "one-size fits all" approach. One critical problem is how to learn the discriminative multi-view characteristics and the unique context of each SCPN.

METHODS

Here, we propose a multi-view coupled self-attention module (MVCS) to capture the global spatial context of the CT image through modeling the association order of space and dimension. Compared with existing self-attention methods, MVCS uses less memory consumption and computational complexity, unearths dimension correlations that previous methods have not found, and is easy to integrate with other frameworks.

RESULTS

In total, a public dataset LUNA16 from LIDC-IDRI, 1319 SCPNs from 1069 patients presenting to a major referral center, and 160 SCPNs from 137 patients from three other major centers were analyzed to pre-train, train, and validate the model. Experimental results showed that performance outperforms the state-of-the-art models in terms of accuracy and stability and is comparable to that of human experts in classifying precancerous lesions and invasive adenocarcinoma. We also provide a fusion MVCS network (MVCSN) by combining the CT image with the clinical characteristics and radiographic features of patients.

CONCLUSION

This tool may ultimately aid in expediting resection of the malignant SCPNs and avoid over-diagnosis of the benign ones, resulting in improved management outcomes.

CLINICAL RELEVANCE STATEMENT

In the diagnosis of sub-centimeter lung adenocarcinoma, fusion MVCSN can help doctors improve work efficiency and guide their treatment decisions to a certain extent.

KEY POINTS

• Advances in computed tomography (CT) not only increase the number of nodules detected, but also the nodules that are identified are smaller, such as sub-centimeter pulmonary nodules (SCPNs). • We propose a multi-view coupled self-attention module (MVCS), which could model spatial and dimensional correlations sequentially for learning global spatial contexts, which is better than other attention mechanisms. • MVCS uses fewer huge memory consumption and computational complexity than the existing self-attention methods when dealing with 3D medical image data. Additionally, it reaches promising accuracy for SCPNs' malignancy evaluation and has lower training cost than other models.

[The contribution of artificial intelligence (AI) subsequent to the processing of thoracic imaging]

The contribution of artificial intelligence (AI) to medical imaging is currently the object of widespread experimentation. The development of deep learning (DL) methods, particularly convolution neural networks (CNNs), has led to performance gains often superior to those achieved by conventional methods such as machine learning. Radiomics is an approach aimed at extracting quantitative data not accessible to the human eye from images expressing a disease. The data subsequently feed machine learning models and produce diagnostic or prognostic probabilities. As for the multiple applications of AI methods in thoracic imaging, they are undergoing evaluation. Chest radiography is a practically ideal field for the development of DL algorithms able to automatically interpret X-rays. Current algorithms can detect up to 14 different abnormalities present either in isolation or in combination. Chest CT is another area offering numerous AI applications. Various algorithms have been specifically formed and validated for the detection and characterization of pulmonary nodules and pulmonary embolism, as well as segmentation and quantitative analysis of the extent of diffuse lung diseases (emphysema, infectious pneumonias, interstitial lung disease). In addition, the analysis of medical images can be associated with clinical, biological, and functional data (multi-omics analysis), the objective being to construct predictive approaches regarding disease prognosis and response to treatment.

AI/ML advances in non-small cell lung cancer biomarker discovery

Lung cancer is the leading cause of cancer deaths among both men and women, representing approximately 25% of cancer fatalities each year. The treatment landscape for non-small cell lung cancer (NSCLC) is rapidly evolving due to the progress made in biomarker-driven targeted therapies. While advancements in targeted treatments have improved survival rates for NSCLC patients with actionable biomarkers, long-term survival remains low, with an overall 5-year relative survival rate below 20%. Artificial intelligence/machine learning (AI/ML) algorithms have shown promise in biomarker discovery, yet NSCLC-specific studies capturing the clinical challenges targeted and emerging patterns identified using AI/ML approaches are lacking. Here, we employed a text-mining approach and identified 215 studies that reported potential biomarkers of NSCLC using AI/ML algorithms. We catalogued these studies with respect to BEST (Biomarkers, EndpointS, and other Tools) biomarker sub-types and summarized emerging patterns and trends in AI/ML-driven NSCLC biomarker discovery. We anticipate that our comprehensive review will contribute to the current understanding of AI/ML advances in NSCLC biomarker research and provide an important catalogue that may facilitate clinical adoption of AI/ML-derived biomarkers.

A deep learning MRI-based signature may provide risk-stratification strategies for nasopharyngeal carcinoma

OBJECTIVE

As the prognosis of nasopharyngeal carcinoma (NPC) is influenced by various factors, making it difficult for clinical physicians to predict the outcome, the objective of this study was to develop a deep learning-based signature for risk stratification in NPC patients.

METHODS

A total of 293 patients were enrolled in the study and divided into training, validation, and testing groups with a ratio of 7:1:2. MRI scans and corresponding clinical information were collected, and the 3-year disease-free survival (DFS) was chosen as the endpoint. The Res-Net18 algorithm was used to develop two deep learning (DL) models and another solely based on clinical characteristics developed by multivariate cox analysis. The performance of both models was evaluated using the area under the curve (AUC) and the concordance index (C-index). Discriminative performance was assessed using Kaplan-Meier survival analysis.

RESULTS

The deep learning approach identified DL prognostic models. The MRI-based DL model showed significantly better performance compared to the traditional model solely based on clinical characteristics (AUC: 0.8861 vs 0.745, p = 0.04 and C-index: 0.865 vs 0.727, p = 0.03). The survival analysis showed significant survival differences between the risk groups identified by the MRI-based model.

CONCLUSION

Our study highlights the potential of MRI in predicting the prognosis of NPC through DL algorithm. This approach has the potential to become a novel tool for prognosis prediction and can help physicians to develop more valid treatment strategies in the future.

Development and evaluation of an integrated model based on a deep segmentation network and demography-added radiomics algorithm for segmentation and diagnosis of early lung adenocarcinoma

Non-invasive early detection and differentiation grading of lung adenocarcinoma using computed tomography (CT) images are clinically important for both clinicians and patients, including determining the extent of lung resection. However, these are difficult to accomplish using preoperative images, with CT-based diagnoses often being different from postoperative pathologic diagnoses. In this study, we proposed an integrated detection and classification algorithm (IDCal) for diagnosing ground-glass opacity nodules (GGN) using CT images and other patient informatics, and compared its performance with that of other diagnostic modalities. All labeling was confirmed by a thoracic surgeon by referring to the patient's CT image and biopsy report. The detection phase was implemented via a modified FC-DenseNet to contour the lesions as elaborately as possible and secure the reliability of the classification phase for subsequent applications. Then, by integrating radiomics features and other patients' general information, the lesions were dichotomously reclassified into "non-invasive" (atypical adenomatous hyperplasia, adenocarcinoma in situ, and minimally invasive adenocarcinoma) and "invasive" (invasive adenocarcinoma). Data from 168 GGN cases were used to develop the IDCal, which was then validated in 31 independent CT scans. IDCal showed a high accuracy of GGN detection (sensitivity, 0.970; false discovery rate, 0.697) and classification (accuracy, 0.97; f1-score, 0.98; ROAUC, 0.96). In conclusion, the proposed IDCal detects and classifies GGN with excellent performance. Thus, it can be suggested that our multimodal prediction model has high potential as an auxiliary diagnostic tool of GGN to help clinicians.

Application value of a computer-aided diagnosis and management system for the detection of lung nodules

Background

Computer-aided diagnosis (CAD) systems can help reduce radiologists' workload. This study assessed the value of a CAD system for the detection of lung nodules on chest computed tomography (CT) images.

Methods

The study retrospectively analyzed the CT images of patients who underwent routine health checkups between August 2019 and November 2019 at 3 hospitals in China. All images were first assessed by 2 radiologists manually in a blinded manner, which was followed by assessment with the CAD system. The location and classification of the lung nodules were determined. The final diagnosis was made by a panel of experts, including 2 associate chief radiologists and 1 chief radiologist at the radiology department. The sensitivity for nodule detection and false-positive nodules per case were calculated.

Results

A total of 1,002 CT images were included in the study, and the process was completed for 999 images. The sensitivity of the CAD system and manual detection was 90.19% and 49.88% (P<0.001), respectively. Similar sensitivity was observed between manual detection and the CAD system in lung nodules >15 mm (P=0.08). The false-positive nodules per case for the CAD system were 0.30±0.84 and those for manual detection were 0.24±0.68 (P=0.12). The sensitivity of the CAD system was higher than that of the radiologists, but the increase in the false-positive rate was only slight.

Conclusions

In addition to reducing the workload for medical professionals, a CAD system developed using a deep-learning model was highly effective and accurate in detecting lung nodules and did not demonstrate a meaningfully higher the false-positive rate.

Discrimination of ground-glass nodular lung adenocarcinoma pathological subtypes via transfer learning: A multicenter study

BACKGROUND

The surgical approach and prognosis for invasive adenocarcinoma (IAC) and minimally invasive adenocarcinoma (MIA) of the lung differ. However, they both manifest as identical ground-glass nodules (GGNs) in computed tomography images, and no effective method exists to discriminate them.

METHODS

We developed and validated a three-dimensional (3D) deep transfer learning model to discriminate IAC from MIA based on CT images of GGNs. This model uses a 3D medical image pre-training model (MedicalNet) and a fusion model to build a classification network. Transfer learning was utilized for end-to-end predictive modeling of the cohort data of the first center, and the cohort data of the other two centers were used as independent external validation data. This study included 999 lung GGN images of 921 patients pathologically diagnosed with IAC or MIA at three cohort centers.

RESULTS

The predictive performance of the model was assessed using the area under the receiver operating characteristic curve (AUC). The model had high diagnostic efficacy for the training and validation groups (accuracy: 89%, sensitivity: 95%, specificity: 84%, and AUC: 95% in the training group; accuracy: 88%, sensitivity: 84%, specificity: 93%, and AUC: 92% in the internal validation group; accuracy: 83%, sensitivity: 83%, specificity: 83%, and AUC: 89% in one external validation group; accuracy: 78%, sensitivity: 80%, specificity: 77%, and AUC: 82% in the other external validation group).

CONCLUSIONS

Our 3D deep transfer learning model provides a noninvasive, low-cost, rapid, and reproducible method for preoperative prediction of IAC and MIA in lung cancer patients with GGNs. It can help clinicians to choose the optimal surgical strategy and improve the prognosis of patients.

A graph neural network model for the diagnosis of lung adenocarcinoma based on multimodal features and an edge-generation network

Background

Lung cancer is a global disease with high lethality, with early screening being considerably helpful for improving the 5-year survival rate. Multimodality features in early screening imaging are an important part of the prediction for lung adenocarcinoma, and establishing a model for adenocarcinoma diagnosis based on multimodal features is an obvious clinical need. Through our practice and investigation, we found that graph neural networks (GNNs) are excellent platforms for multimodal feature fusion, and the data can be completed using the edge-generation network. Therefore, we propose a new lung adenocarcinoma multiclassification model based on multimodal features and an edge-generation network.

Methods

According to a ratio of 80% to 20%, respectively, the dataset of 338 cases was divided into the training set and the test set through 5-fold cross-validation, and the distribution of the 2 sets was the same. First, the regions of interest (ROIs) cropped from computed tomography (CT) images were separately fed into convolutional neural networks (CNNs) and radiomics processing platforms. The results of the 2 parts were then input into a graph embedding representation network to obtain the fused feature vectors. Subsequently, a graph database based on the clinical and semantic features was established, and the data were supplemented by an edge-generation network, with the fused feature vectors being used as the input of the nodes. This enabled us to clearly understand where the information transmission of the GNN takes place and improves the interpretability of the model. Finally, the nodes were classified using GNNs.

Results

On our dataset, the proposed method presented in this paper achieved superior results compared to traditional methods and showed some comparability with state-of-the-art methods for lung nodule classification. The results of our method are as follows: accuracy (ACC) =66.26% (±4.46%), area under the curve (AUC) =75.86% (±1.79%), F1-score =64.00% (±3.65%), and Matthews correlation coefficient (MCC) =48.40% (±5.07%). The model with the edge-generating network consistently outperformed the model without it in all aspects.

Conclusions

The experiments demonstrate that with appropriate data=construction methods GNNs can outperform traditional image processing methods in the field of CT-based medical image classification. Additionally, our model has higher interpretability, as it employs subjective clinical and semantic features as the data construction approach. This will help doctors better leverage human-computer interactions.

Neural network-based model for evaluating inert nodules and volume doubling time in T1 lung adenocarcinoma: a nested case-control study

Objective

The purpose of this study is to establish model for assessing inert nodules predicting nodule volume-doubling.

Methods

A total of 201 patients with T1 lung adenocarcinoma were analysed retrospectively pulmonary nodule information was predicted by an AI pulmonary nodule auxiliary diagnosis system. The nodules were classified into two groups: inert nodules (volume-doubling time (VDT)>600 days n=152) noninert nodules (VDT<600 days n=49). Then taking the clinical imaging features obtained at the first examination as predictive variables the inert nodule judgement model <sn</sn>>(INM) volume-doubling time estimation model (VDTM) were constructed based on a deep learning-based neural network. The performance of the INM was evaluated by the area under the curve (AUC) obtained from receiver operating characteristic (ROC) analysis the performance of the VDTM was evaluated by R²(determination coefficient).

Results

The accuracy of the INM in the training and testing cohorts was 81.13% and 77.50%, respectively. The AUC of the INM in the training and testing cohorts was 0.7707 (95% CI 0.6779-0.8636) and 0.7700 (95% CI 0.5988-0.9412), respectively. The INM was effective in identifying inert pulmonary nodules; additionally, the R2 of the VDTM in the training cohort was 0.8008, and that in the testing cohort was 0.6268. The VDTM showed moderate performance in estimating the VDT, which can provide some reference during a patients' first examination and consultation.

Conclusion

The INM and the VDTM based on deep learning can help radiologists and clinicians distinguish among inert nodules and predict the nodule volume-doubling time to accurately treat patients with pulmonary nodules.

The use of the mean computed-tomography value to predict the invasiveness of ground-glass nodules: A meta-analysis

The invasiveness of ground-glass nodules (GGNs) is difficult to characterize through morphological examination. Multiple studies have independently detected a close relationship between mean computed tomography value and invasiveness of GGNs, however, their relative diagnostic accuracy is uncertain. Here, we performed a meta-analysis to validate whether the mean computed tomography value can predict the invasiveness of GGNs. Briefly, we searched the Web of Science, Embase, PubMed, Cochrane, Google Scholar, CNKI, VIP, Wanfang and SinoMed databases. The sensitivity, specificity, 95% confidence interval (CI), symmetric receiver operating characteristic curve (SROC curve) and the area under curve (AUC) were obtained using STATA 16.0 to evaluate the predictive value of the mean computed tomography value for GGNs. The presence of heterogeneity was assessed using fixed effects sensitivity analysis and I² statistics. We used the Deek's funnel plot to evaluate the possibility of publication bias. Thirteen studies encompassing 1564 GGNs were included in our meta-analysis. Six of these studies revealed that using the mean computed tomography value for the diagnosis of pre-invasive and invasive lesions had a sensitivity and specificity of 0.75 (95% CI: 0.61-0.85) and 0.81 (95% CI: 0.74-0.86), respectively. The optimal critical value was -557 Hu. Later, eight studies were examined for the use of the mean CT value for patients with minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC); the results showed that the sensitivity was 0.78 (95% CI: 0.66-0.86) and the specificity was 0.81 (95% CI: 0.68-0.89), and the optimal critical value was -484 Hu. Therefore, the mean computed tomography value assessed via CT scan could be a significant predictor of the invasiveness of GGNs as well as a good surgical treatment guide in patients diagnosed with lung cancer. PROSPERO REGISTRATION NUMBER: CRD42020177125.

Feature selection methods and predictive models in CT lung cancer radiomics

Radiomics is a technique that extracts quantitative features from medical images using data-characterization algorithms. Radiomic features can be used to identify tissue characteristics and radiologic phenotyping that is not observable by clinicians. A typical workflow for a radiomics study includes cohort selection, radiomic feature extraction, feature and predictive model selection, and model training and validation. While there has been increasing attention given to radiomic feature extraction, standardization, and reproducibility, currently, there is a lack of rigorous evaluation of feature selection methods and predictive models. Herein, we review the published radiomics investigations in CT lung cancer and provide an overview of the commonly used radiomic feature selection methods and predictive models. We also compare limitations of various methods in clinical applications and present sources of uncertainty associated with those methods. This review is expected to help raise awareness of the impact of radiomic feature and model selection methods on the integrity of radiomics studies.

Artificial intelligence in lung cancer: current applications and perspectives

Artificial intelligence (AI) has been a very active research topic over the last years and thoracic imaging has particularly benefited from the development of AI and in particular deep learning. We have now entered a phase of adopting AI into clinical practice. The objective of this article was to review the current applications and perspectives of AI in thoracic oncology. For pulmonary nodule detection, computer-aided detection (CADe) tools have been commercially available since the early 2000s. The more recent rise of deep learning and the availability of large annotated lung nodule datasets have allowed the development of new CADe tools with fewer false-positive results per examination. Classical machine learning and deep-learning methods were also used for pulmonary nodule segmentation allowing nodule volumetry and pulmonary nodule characterization. For pulmonary nodule characterization, radiomics and deep-learning approaches were used. Data from the National Lung Cancer Screening Trial (NLST) allowed the development of several computer-aided diagnostic (CADx) tools for diagnosing lung cancer on chest computed tomography. Finally, AI has been used as a means to perform virtual biopsies and to predict response to treatment or survival. Thus, many detection, characterization and stratification tools have been proposed, some of which are commercially available.

How does the artificial intelligence-based image-assisted technique help physicians in diagnosis of pulmonary adenocarcinoma? A randomized controlled experiment of multicenter physicians in China

OBJECTIVE

Although artificial intelligence (AI) has achieved high levels of accuracy in the diagnosis of various diseases, its impact on physicians' decision-making performance in clinical practice is uncertain. This study aims to assess the impact of AI on the diagnostic performance of physicians with differing levels of self-efficacy under working conditions involving different time pressures.

MATERIALS AND METHODS

A 2 (independent diagnosis vs AI-assisted diagnosis) × 2 (no time pressure vs 2-minute time limit) randomized controlled experiment of multicenter physicians was conducted. Participants diagnosed 10 pulmonary adenocarcinoma cases and their diagnostic accuracy, sensitivity, and specificity were evaluated. Data analysis was performed using multilevel logistic regression.

RESULTS

One hundred and four radiologists from 102 hospitals completed the experiment. The results reveal (1) AI greatly increases physicians' diagnostic accuracy, either with or without time pressure; (2) when no time pressure, AI significantly improves physicians' diagnostic sensitivity but no significant change in specificity, while under time pressure, physicians' diagnostic sensitivity and specificity are both improved with the aid of AI; (3) when no time pressure, physicians with low self-efficacy benefit from AI assistance thus improving diagnostic accuracy but those with high self-efficacy do not, whereas physicians with low and high levels of self-efficacy both benefit from AI under time pressure.

DISCUSSION

This study is one of the first to provide real-world evidence regarding the impact of AI on physicians' decision-making performance, taking into account 2 boundary factors: clinical time pressure and physicians' self-efficacy.

CONCLUSION

AI-assisted diagnosis should be prioritized for physicians working under time pressure or with low self-efficacy.

Thoracic Imaging in China: Yesterday, Today, and Tomorrow

Thoracic imaging has been revolutionized through advances in technology and research around the world, and so has China. Thoracic imaging in China has progressed from anatomic observation to quantitative and functional evaluation, from using traditional approaches to using artificial intelligence. This article will review the past, present, and future of thoracic imaging in China, in an attempt to establish new accepted strategies moving forward.

3D deep learning versus the current methods for predicting tumor invasiveness of lung adenocarcinoma based on high-resolution computed tomography images

Background

Different pathological subtypes of lung adenocarcinoma lead to different treatment decisions and prognoses, and it is clinically important to distinguish invasive lung adenocarcinoma from preinvasive adenocarcinoma (adenocarcinoma in situ and minimally invasive adenocarcinoma). This study aims to investigate the performance of the deep learning approach based on high-resolution computed tomography (HRCT) images in the classification of tumor invasiveness and compare it with the performances of currently available approaches.

Methods

In this study, we used a deep learning approach based on 3D conventional networks to automatically predict the invasiveness of pulmonary nodules. A total of 901 early-stage non-small cell lung cancer patients who underwent surgical treatment at Shanghai Chest Hospital between November 2015 and March 2017 were retrospectively included and randomly assigned to a training set (n=814) or testing set 1 (n=87). We subsequently included 116 patients who underwent surgical treatment and intraoperative frozen section between April 2019 and January 2020 to form testing set 2. We compared the performance of our deep learning approach in predicting tumor invasiveness with that of intraoperative frozen section analysis and human experts (radiologists and surgeons).

Results

The deep learning approach yielded an area under the receiver operating characteristic curve (AUC) of 0.946 for distinguishing preinvasive adenocarcinoma from invasive lung adenocarcinoma in the testing set 1, which is significantly higher than the AUCs of human experts (P<0.05). In testing set 2, the deep learning approach distinguished invasive adenocarcinoma from preinvasive adenocarcinoma with an AUC of 0.862, which is higher than that of frozen section analysis (0.755, P=0.043), senior thoracic surgeons (0.720, P=0.006), radiologists (0.766, P>0.05) and junior thoracic surgeons (0.768, P>0.05).

Conclusions

We developed a deep learning model that achieved comparable performance to intraoperative frozen section analysis in determining tumor invasiveness. The proposed method may contribute to clinical decisions related to the extent of surgical resection.

ViSTA: A Novel Network Improving Lung Adenocarcinoma Invasiveness Prediction from Follow-Up CT Series

Simple Summary

Assessing follow-up computed tomography(CT) series is of great importance in clinical practice for lung nodule diagnosis. Deep learning is a thriving data mining method in medical imaging and has obtained surprising results. However, previous studies mostly focused on the analysis of single static time points instead of the entire follow-up series and required regular intervals between CT examinations. In the current study, we propose a new deep learning framework, named ViSTA, that can better evaluate tumor invasiveness using irregularly serial follow-up CT images to avoid aggressive procedures or delay diagnosis in clinical practice. ViSTA provides a new solution for irregularly sampled data. ViSTA delivers superior performance compared with other static or serial deep learning models. The proposed ViSTA framework is capable of improving performance close to the human level in the prediction of invasiveness of lung adenocarcinoma while being transferrable to other tasks analyzing serial medical data.

Abstract

To investigate the value of the deep learning method in predicting the invasiveness of early lung adenocarcinoma based on irregularly sampled follow-up computed tomography (CT) scans. In total, 351 nodules were enrolled in the study. A new deep learning network based on temporal attention, named Visual Simple Temporal Attention (ViSTA), was proposed to process irregularly sampled follow-up CT scans. We conducted substantial experiments to investigate the supplemental value in predicting the invasiveness using serial CTs. A test set composed of 69 lung nodules was reviewed by three radiologists. The performance of the model and radiologists were compared and analyzed. We also performed a visual investigation to explore the inherent growth pattern of the early adenocarcinomas. Among counterpart models, ViSTA showed the best performance (AUC: 86.4% vs. 60.6%, 75.9%, 66.9%, 73.9%, 76.5%, 78.3%). ViSTA also outperformed the model based on Volume Doubling Time (AUC: 60.6%). ViSTA scored higher than two junior radiologists (accuracy of 81.2% vs. 75.4% and 71.0%) and came close to the senior radiologist (85.5%). Our proposed model using irregularly sampled follow-up CT scans achieved promising accuracy in evaluating the invasiveness of the early stage lung adenocarcinoma. Its performance is comparable with senior experts and better than junior experts and traditional deep learning models. With further validation, it can potentially be applied in clinical practice.

Lung cancer CT image generation from a free-form sketch using style-based pix2pix for data augmentation

Artificial intelligence (AI) applications in medical imaging continue facing the difficulty in collecting and using large datasets. One method proposed for solving this problem is data augmentation using fictitious images generated by generative adversarial networks (GANs). However, applying a GAN as a data augmentation technique has not been explored, owing to the quality and diversity of the generated images. To promote such applications by generating diverse images, this study aims to generate free-form lesion images from tumor sketches using a pix2pix-based model, which is an image-to-image translation model derived from GAN. As pix2pix, which assumes one-to-one image generation, is unsuitable for data augmentation, we propose StylePix2pix, which is independently improved to allow one-to-many image generation. The proposed model introduces a mapping network and style blocks from StyleGAN. Image generation results based on 20 tumor sketches created by a physician demonstrated that the proposed method can reproduce tumors with complex shapes. Additionally, the one-to-many image generation of StylePix2pix suggests effectiveness in data-augmentation applications.

Development, Validation, and Comparison of Image-Based, Clinical Feature-Based and Fusion Artificial Intelligence Diagnostic Models in Differentiating Benign and Malignant Pulmonary Ground-Glass Nodules

Objective

This study aimed to develop effective artificial intelligence (AI) diagnostic models based on CT images of pulmonary nodules only, on descriptional and quantitative clinical or image features, or on a combination of both to differentiate benign and malignant ground-glass nodules (GGNs) to assist in the determination of surgical intervention.

Methods

Our study included a total of 867 nodules (benign nodules: 112; malignant nodules: 755) with postoperative pathological diagnoses from two centers. For the diagnostic models to discriminate between benign and malignant GGNs, we adopted three different artificial intelligence (AI) approaches: a) an image-based deep learning approach to build a deep neural network (DNN); b) a clinical feature-based machine learning approach based on the clinical and image features of nodules; c) a fusion diagnostic model integrating the original images and the clinical and image features. The performance of the models was evaluated on an internal test dataset (the “Changzheng Dataset”) and an independent test dataset collected from an external institute (the “Longyan Dataset”). In addition, the performance of automatic diagnostic models was compared with that of manual evaluations by two radiologists on the ‘Longyan dataset’.

Results

The image-based deep learning model achieved an appealing diagnostic performance, yielding AUC values of 0.75 (95% confidence interval [CI]: 0.62, 0.89) and 0.76 (95% CI: 0.61, 0.90), respectively, on both the Changzheng and Longyan datasets. The clinical feature-based machine learning model performed well on the Changzheng dataset (AUC, 0.80 [95% CI: 0.64, 0.96]), whereas it performed poorly on the Longyan dataset (AUC, 0.62 [95% CI: 0.42, 0.83]). The fusion diagnostic model achieved the best performance on both the Changzheng dataset (AUC, 0.82 [95% CI: 0.71-0.93]) and the Longyan dataset (AUC, 0.83 [95% CI: 0.70-0.96]), and it achieved a better specificity (0.69) than the radiologists (0.33-0.44) on the Longyan dataset.

Conclusion

The deep learning models, including both the image-based deep learning model and the fusion model, have the ability to assist radiologists in differentiating between benign and malignant nodules for the precise management of patients with GGNs.

A narrative review of deep learning applications in lung cancer research: from screening to prognostication

Background and Objective

Deep learning (DL) algorithms have been developed for various tasks, including lung nodule detection on chest radiographs or lung cancer computed tomography screening, potential candidate selection in lung cancer screening, malignancy prediction for indeterminate pulmonary nodules, lung cancer staging, treatment response prediction, prognostication, and prediction of genetic mutations in lung cancer. Furthermore, these DL algorithms have been applied in various clinical settings in order for them to be generalized in real-world clinical practice. Multiple DL algorithms have been corroborated to be on par with experts or current clinical prediction models for several specific tasks. However, no article has yet comprehensively reviewed DL algorithms dedicated to lung cancer research. This narrative review presents an overview of the literature dealing with DL techniques applied in lung cancer research and briefly summarizes the results according to the DL algorithms’ clinical use cases.

Methods

we performed a narrative review by searching the Embase and OVID-MEDLINE databases for articles published in English from October, 2016 until September, 2021 and reviewing the bibliographies of key references to identify important literature related to DL in lung cancer research. The background, development, results, and clinical implications of each DL algorithm are briefly discussed. Lastly, we end this review article by highlighting future directions in lung cancer research using DL techniques.

Key Content and Findings

DL algorithms have been introduced to show comparable or higher performance than human experts in various clinical settings. Specifically, they have been actively applied to detect lung nodules in chest radiographs or computed tomography (CT) examinations, optimize candidate selection for lung cancer screening (LCS), predict the malignancy of lung nodules, stage lung cancer, and predict treatment response, patients’ prognoses, and genetic mutations in lung cancers.

Conclusions

DL algorithms have corroborated their potential value for various tasks, ranging from lung cancer screening to prognostication of lung cancer patients. Future research is warranted for the clinical application of these algorithms in daily clinical practice and verification of their real-world clinical usefulness.

Predictive Efficacy of a Radiomics Random Forest Model for Identifying Pathological Subtypes of Lung Adenocarcinoma Presenting as Ground-Glass Nodules

Purpose

To establish and verify the ability of a radiomics prediction model to distinguish invasive adenocarcinoma (IAC) and minimal invasive adenocarcinoma (MIA) presenting as ground-glass nodules (GGNs).

Methods

We retrospectively analyzed 118 lung GGN images and clinical data from 106 patients in our hospital from March 2016 to April 2019. All pathological classifications of lung GGN were confirmed as IAC or MIA by two pathologists. R language software (version 3.5.1) was used for the statistical analysis of the general clinical data. ITK-SNAP (version 3.6) and A.K. software (Analysis Kit, American GE Company) were used to manually outline the regions of interest of lung GGNs and collect three-dimensional radiomics features. Patients were randomly divided into training and verification groups (ratio, 7:3). Random forest combined with hyperparameter tuning was used for feature selection and prediction modeling. The receiver operating characteristic curve and the area under the curve (AUC) were used to evaluate model prediction efficacy. The calibration curve was used to evaluate the calibration effect.

Results

There was no significant difference between IAC and MIA in terms of age, gender, smoking history, tumor history, and lung GGN location in both the training and verification groups (P>0.05). For each lung GGN, the collected data included 396 three-dimensional radiomics features in six categories. Based on the training cohort, nine optimal radiomics features in three categories were finally screened out, and a prediction model was established. We found that the training group had a high diagnostic efficacy [accuracy, sensitivity, specificity, and AUC of the training group were 0.89 (95%CI, 0.73 - 0.99), 0.98 (95%CI, 0.78 - 1.00), 0.81 (95%CI, 0.59 - 1.00), and 0.97 (95%CI, 0.92-1.00), respectively; those of the validation group were 0.80 (95%CI, 0.58 - 0.93), 0.82 (95%CI, 0.55 - 1.00), 0.78 (95%CI, 0.57 - 1.00), and 0.92 (95%CI, 0.83 - 1.00), respectively]. The model calibration curve showed good consistency between the predicted and actual probabilities.

Conclusions

The radiomics prediction model established by combining random forest with hyperparameter tuning effectively distinguished IAC from MIA presenting as GGNs and represents a noninvasive, low-cost, rapid, and reproducible preoperative prediction method for clinical application.

Multitask Learning Radiomics on Longitudinal Imaging to Predict Survival Outcomes following Risk-Adaptive Chemoradiation for Non-Small Cell Lung Cancer

Medical imaging provides quantitative and spatial information to evaluate treatment response in the management of patients with non-small cell lung cancer (NSCLC). High throughput extraction of radiomic features on these images can potentially phenotype tumors non-invasively and support risk stratification based on survival outcome prediction. The prognostic value of radiomics from different imaging modalities and time points prior to and during chemoradiation therapy of NSCLC, relative to conventional imaging biomarker or delta radiomics models, remains uncharacterized. We investigated the utility of multitask learning of multi-time point radiomic features, as opposed to single-task learning, for improving survival outcome prediction relative to conventional clinical imaging feature model benchmarks. Survival outcomes were prospectively collected for 45 patients with unresectable NSCLC enrolled on the FLARE-RT phase II trial of risk-adaptive chemoradiation and optional consolidation PD-L1 checkpoint blockade (NCT02773238). FDG-PET, CT, and perfusion SPECT imaging pretreatment and week 3 mid-treatment was performed and 110 IBSI-compliant pyradiomics shape-/intensity-/texture-based features from the metabolic tumor volume were extracted. Outcome modeling consisted of a fused Laplacian sparse group LASSO with component-wise gradient boosting survival regression in a multitask learning framework. Testing performance under stratified 10-fold cross-validation was evaluated for multitask learning radiomics of different imaging modalities and time points. Multitask learning models were benchmarked against conventional clinical imaging and delta radiomics models and evaluated with the concordance index (c-index) and index of prediction accuracy (IPA). FDG-PET radiomics had higher prognostic value for overall survival in test folds (c-index 0.71 [0.67, 0.75]) than CT radiomics (c-index 0.64 [0.60, 0.71]) or perfusion SPECT radiomics (c-index 0.60 [0.57, 0.63]). Multitask learning of pre-/mid-treatment FDG-PET radiomics (c-index 0.71 [0.67, 0.75]) outperformed benchmark clinical imaging (c-index 0.65 [0.59, 0.71]) and FDG-PET delta radiomics (c-index 0.52 [0.48, 0.58]) models. Similarly, the IPA for multitask learning FDG-PET radiomics (30%) was higher than clinical imaging (26%) and delta radiomics (15%) models. Radiomics models performed consistently under different voxel resampling conditions. Multitask learning radiomics for outcome modeling provides a clinical decision support platform that leverages longitudinal imaging information. This framework can reveal the relative importance of different imaging modalities and time points when designing risk-adaptive cancer treatment strategies.

IMAL-Net: Interpretable multi-task attention learning network for invasive lung adenocarcinoma screening in CT images

PURPOSE

Feature maps created from deep convolutional neural networks (DCNNs) have been widely used for visual explanation of DCNN-based classification tasks. However, many clinical applications such as benign-malignant classification of lung nodules normally require quantitative and objective interpretability, rather than just visualization. In this paper, we propose a novel interpretable multi-task attention learning network named IMAL-Net for early invasive adenocarcinoma screening in chest computed tomography images, which takes advantage of segmentation prior to assist interpretable classification.

METHODS

Two sub-ResNets are firstly integrated together via a prior-attention mechanism for simultaneous nodule segmentation and invasiveness classification. Then, numerous radiomic features from the segmentation results are concatenated with high-level semantic features from the classification subnetwork by FC layers to achieve superior performance. Meanwhile, an end-to-end feature selection mechanism (named FSM) is designed to quantify crucial radiomic features greatly affecting the prediction of each sample, and thus it can provide clinically applicable interpretability to the prediction result.

RESULTS

Nodule samples from a total of 1626 patients were collected from two grade-A hospitals for large-scale verification. Five-fold cross validation demonstrated that the proposed IMAL-Net can achieve an AUC score of 93.8% ± 1.1% and a recall score of 93.8% ± 2.8% for identification of invasive lung adenocarcinoma.

CONCLUSIONS

It can be concluded that fusing semantic features and radiomic features can achieve obvious improvements in the invasiveness classification task. Moreover, by learning more fine-grained semantic features and highlighting the most important radiomics features, the proposed attention and FSM mechanisms not only can further improve the performance but also can be used for both visual explanations and objective analysis of the classification results.

Consecutive Serial Non-Contrast CT Scan-Based Deep Learning Model Facilitates the Prediction of Tumor Invasiveness of Ground-Glass Nodules

Introduction

Tumors are continuously evolving biological systems which can be monitored by medical imaging. Previous studies only focus on single timepoint images, whether the performance could be further improved by using serial noncontrast CT imaging obtained during nodule follow-up management remains unclear. In this study, we evaluated DL model for predicting tumor invasiveness of GGNs through analyzing time series CT images.

Methods

A total of 168 pathologically confirmed GGN cases (48 noninvasive lesions and 120 invasive lesions) were retrospectively collected and randomly assigned to the development dataset (n = 123) and independent testing dataset (n = 45). All patients underwent consecutive noncontrast CT examinations, and the baseline CT and 3-month follow-up CT images were collected. The gross region of interest (ROI) patches containing only tumor region and the full ROI patches including both tumor and peritumor regions were cropped from CT images. A baseline model was built on the image features and demographic features. Four DL models were proposed: two single-DL model using gross ROI (model 1) or full ROI patches (model 3) from baseline CT images, and two serial-DL models using gross ROI (model 2) or full ROI patches (model 4) from consecutive CT images (baseline scan and 3-month follow-up scan). In addition, a combined model integrating serial full ROI patches and clinical information was also constructed. The performance of these predictive models was assessed with respect to discrimination and clinical usefulness.

Results

The area under the curve (AUC) of the baseline model, models 1, 2, 3, and 4 were 0.562 [(95% confidence interval (C)], 0.406~0.710), 0.693 (95% CI, 0.538-0.822), 0.787 (95% CI, 0.639-0.895), 0.727 (95% CI, 0.573-0.849), and 0.811 (95% CI, 0.667-0.912) in the independent testing dataset, respectively. The results indicated that the peritumor region had potential to contribute to tumor invasiveness prediction, and the model performance was further improved by integrating imaging scans at multiple timepoints. Furthermore, the combined model showed best discrimination ability, with AUC, sensitivity, specificity, and accuracy achieving 0.831 (95% CI, 0.690-0.926), 86.7%, 73.3%, and 82.2%, respectively.

Conclusion

The DL model integrating full ROIs from serial CT images shows improved predictive performance in differentiating noninvasive from invasive GGNs than the model using only baseline CT images, which could benefit the clinical management of GGNs.

CT-based radiomics for differentiating invasive adenocarcinomas from indolent lung adenocarcinomas appearing as ground-glass nodules: Asystematic review

PURPOSE

To provide an overview of the available studies investigating the use of computer tomography (CT) radiomics features for differentiating invasive adenocarcinomas (IAC) from indolent lung adenocarcinomas presenting as ground-glass nodules (GGNs), to identify the bias of the studies and to propose directions for future research.

METHOD

PubMed, Embase, Web of Science Core Collection were searched for relevant studies. The studies differentiating IAC from indolent lung adenocarcinomas appearing as GGNs based on CT radiomics features were included. Basic information, patient information, CT-scanner information, technique information and performance information were extracted for each included study. The quality of each study was assessed using the Radiomic Quality Score (RQS) and the Prediction model Risk of Bias Assessment Tool (PROBAST).

RESULTS

Twenty-eight studies were included with patients ranging from 34 to 794. All of them were retrospective. Patients in three studies were from multiple centers. Most studies segmented regions of interest manually. Pyradiomics and AK software were the most frequently used for features extraction. The number of radiomics features extracted varied from 7 to 10329. Logistic regression was the most frequently chosen model. Entropy was identified as radiomics signature in seven studies. The AUC of included studies ranged from 0.77 to 0.98 in 15 validation sets. The percentage RQS ranged from 3% to 50%. According to PROBAST, the overall risk of bias (ROB) was high in 89.3% (25/28) of included studies, unclear in 7.1% (2/28) of included studies, and low in 3.6% (1/28) of included studies. All studies were low concern regarding the applicability of primary studies to the review question.

CONCLUSION

CT radiomics-based model is promising and encouraging in differentiating IAC from indolent lung adenocarcinomas, though they require methodological rigor. Well-designed studies are necessary to demonstrate their validity and standardization of methods and results can prompt their use in daily clinical practice.

A deep learning-based radiomic nomogram for prognosis and treatment decision in advanced nasopharyngeal carcinoma: A multicentre study

Background

Induction chemotherapy (ICT) plus concurrent chemoradiotherapy (CCRT) and CCRT alone were the optional treatment regimens in locoregionally advanced nasopharyngeal carcinoma (NPC) patients. Currently, the choice of them remains equivocal in clinical practice. We aimed to develop a deep learning-based model for treatment decision in NPC.

Methods

A total of 1872 patients with stage T3N1M0 NPC were enrolled from four Chinese centres and received either ICT+CCRT or CCRT. A nomogram was constructed for predicting the prognosis of patients with different treatment regimens using multi-task deep learning radiomics and pre-treatment MR images, based on which an optimal treatment regimen was recommended. Model performance was assessed by the concordance index (C-index) and the Kaplan-Meier estimator.

Findings

The nomogram showed excellent prognostic ability for disease-free survival in both the CCRT (C-index range: 0.888-0.921) and ICT+CCRT (C-index range: 0.784-0.830) groups. According to the prognostic difference between treatments using the nomogram, patients were divided into the ICT-preferred and CCRT-preferred groups. In the ICT-preferred group, patients receiving ICT+CCRT exhibited prolonged survival over those receiving CCRT in the internal and external test cohorts (hazard ratio [HR]: 0.17, p<0.001 and 0.24, p=0.02); while the trend was opposite in the CCRT-preferred group (HR: 6.24, p<0.001 and 12.08, p<0.001). Similar results for treatment decision using the nomogram were obtained in different subgroups stratified by clinical factors and MR acquisition parameters.

Interpretation

Our nomogram could predict the prognosis of T3N1M0 NPC patients with different treatment regimens and accordingly recommend an optimal treatment regimen, which may serve as a potential tool for promoting personalized treatment of NPC.

Funding

National Key R&D Program of China, National Natural Science Foundation of China, Beijing Natural Science Foundation, Strategic Priority Research Program of CAS, Project of High-Level Talents Team Introduction in Zhuhai City, Beijing Natural Science Foundation, Beijing Nova Program, Youth Innovation Promotion Association CAS.

Deep learning for predicting subtype classification and survival of lung adenocarcinoma on computed tomography

OBJECTIVES

The subtype classification of lung adenocarcinoma is important for treatment decision. This study aimed to investigate the deep learning and radiomics networks for predicting histologic subtype classification and survival of lung adenocarcinoma diagnosed through computed tomography (CT) images.

METHODS

A dataset of 1222 patients with lung adenocarcinoma were retrospectively enrolled from three medical institutions. The anonymised preoperative CT images and pathological labels of atypical adenomatous hyperplasia, adenocarcinoma in situ, minimally invasive adenocarcinoma, invasive adenocarcinoma (IAC) with five predominant components were obtained. These pathological labels were divided into 2-category classification (IAC; non-IAC), 3-category and 8-category. We modeled the classification task of histological subtypes based on modified ResNet-34 deep learning network, radiomics strategies and deep radiomics combined algorithm. Then we established the prognostic models in lung adenocarcinoma patients with survival outcomes. The accuracy (ACC), area under ROC curves (AUCs) and C-index were primarily performed to evaluate the algorithms.

RESULTS

This study included a training set (n = 802) and two validation cohorts (internal, n = 196; external, n = 224). The ACC of deep radiomics algorithm in internal validation achieved 0.8776, 0.8061 in the 2-category, 3-category classification, respectively. Even in 8 classifications, the AUC ranged from 0.739 to 0.940 in internal set. Further, we constructed a prognosis model that C-index was 0.892(95% CI: 0.846-0.937) in internal validation set.

CONCLUSIONS

The automated deep radiomics based triage system has achieved the great performance in the subtype classification and survival predictability in patients with CT-detected lung adenocarcinoma nodules, providing the clinical guide for treatment strategies.

Deep Learning-Based Stage-Wise Risk Stratification for Early Lung Adenocarcinoma in CT Images: A Multi-Center Study

Simple Summary

Prediction of the malignancy and invasiveness of ground glass nodules (GGNs) from computed tomography images is a crucial task for radiologists in risk stratification of early-stage lung adenocarcinoma. In order to solve this challenge, a two-stage deep neural network (DNN) was developed based on the images collected from four centers. A multi-reader multi-case observer study was conducted to evaluate the model capability. The performance of our model was comparable or even more accurate than that of senior radiologists, with average area under the curve values of 0.76 and 0.95 for two tasks, respectively. Findings suggest (1) a positive trend between the diagnostic performance and radiologist’s experience, (2) DNN yielded equivalent or even higher performance in comparison with senior radiologists, and (3) low image resolution reduced the model performance in predicting the risks of GGNs.

Abstract

This study aims to develop a deep neural network (DNN)-based two-stage risk stratification model for early lung adenocarcinomas in CT images, and investigate the performance compared with practicing radiologists. A total of 2393 GGNs were retrospectively collected from 2105 patients in four centers. All the pathologic results of GGNs were obtained from surgically resected specimens. A two-stage deep neural network was developed based on the 3D residual network and atrous convolution module to diagnose benign and malignant GGNs (Task1) and classify between invasive adenocarcinoma (IA) and non-IA for these malignant GGNs (Task2). A multi-reader multi-case observer study with six board-certified radiologists’ (average experience 11 years, range 2–28 years) participation was conducted to evaluate the model capability. DNN yielded area under the receiver operating characteristic curve (AUC) values of 0.76 ± 0.03 (95% confidence interval (CI): (0.69, 0.82)) and 0.96 ± 0.02 (95% CI: (0.92, 0.98)) for Task1 and Task2, which were equivalent to or higher than radiologists in the senior group with average AUC values of 0.76 and 0.95, respectively (p > 0.05). With the CT image slice thickness increasing from 1.15 mm ± 0.36 to 1.73 mm ± 0.64, DNN performance decreased 0.08 and 0.22 for the two tasks. The results demonstrated (1) a positive trend between the diagnostic performance and radiologist’s experience, (2) the DNN yielded equivalent or even higher performance in comparison with senior radiologists, and (3) low image resolution decreased model performance in predicting the risks of GGNs. Once tested prospectively in clinical practice, the DNN could have the potential to assist doctors in precision diagnosis and treatment of early lung adenocarcinoma.

Artificial intelligence-based analysis for immunohistochemistry staining of immune checkpoints to predict resected non-small cell lung cancer survival and relapse

Background

Conventional analysis of single-plex chromogenic immunohistochemistry (IHC) focused on quantitative but spatial analysis. How immune checkpoints localization related to non-small cell lung cancer (NSCLC) prognosis remained unclear.

Methods

Here, we analyzed ten immune checkpoints on 1,859 tumor microarrays (TMAs) from 121 NSCLC patients and recruited an external cohort of 30 NSCLC patients with 214 whole-slide IHC. EfficientUnet was applied to segment tumor cells (TCs) and tumor-infiltrating lymphocytes (TILs), while ResNet was performed to extract prognostic features from IHC images.

Results

The features of galectin-9, OX40, OX40L, KIR2D, and KIR3D played an un-negatable contribution to overall survival (OS) and relapse-free survival (RFS) in the internal cohort, validated in public databases (GEPIA, HPA, and STRING). The IC-Score and Res-Score were two predictive models established by EfficientUnet and ResNet. Based on the IC-Score, Res-Score, and clinical features, the integrated score presented the highest AUC for OS and RFS, which could achieve 0.9 and 0.85 in the internal testing cohort. The robustness of Res-Score was validated in the external cohort (AUC: 0.80–0.87 for OS, and 0.83–0.94 for RFS). Additionally, the neutrophil-to-lymphocyte ratio (NLR) combined with the PD-1/PD-L1 signature established by EfficientUnet can be a predictor for RFS in the external cohort.

Conclusions

Overall, we established a reliable model to risk-stratify relapse and death in NSCLC with a generalization ability, which provided a convenient approach to spatial analysis of single-plex chromogenic IHC.

Structural and functional radiomics for lung cancer

INTRODUCTION

Lung cancer ranks second in new cancer cases and first in cancer-related deaths worldwide. Precision medicine is working on altering treatment approaches and improving outcomes in this patient population. Radiological images are a powerful non-invasive tool in the screening and diagnosis of early-stage lung cancer, treatment strategy support, prognosis assessment, and follow-up for advanced-stage lung cancer. Recently, radiological features have evolved from solely semantic to include (handcrafted and deep) radiomic features. Radiomics entails the extraction and analysis of quantitative features from medical images using mathematical and machine learning methods to explore possible ties with biology and clinical outcomes.

METHODS

Here, we outline the latest applications of both structural and functional radiomics in detection, diagnosis, and prediction of pathology, gene mutation, treatment strategy, follow-up, treatment response evaluation, and prognosis in the field of lung cancer.

CONCLUSION

The major drawbacks of radiomics are the lack of large datasets with high-quality data, standardization of methodology, the black-box nature of deep learning, and reproducibility. The prerequisite for the clinical implementation of radiomics is that these limitations are addressed. Future directions include a safer and more efficient model-training mode, merge multi-modality images, and combined multi-discipline or multi-omics to form "Medomics."