A causality-inspired generalized model for automated pancreatic cancer diagnosis.

Pancreatic cancer (PC) is a severely malignant cancer variant with high mortality. Since PC has no obvious symptoms, most PC patients are belatedly diagnosed at advanced disease stages. Recently, artificial intelligence (AI) approaches have demonstrated promising prospects for early diagnosis of pancreatic cancer. However, certain non-causal factors (such as intensity and texture appearance variations, also called confounders) tend to induce spurious correlation with PC diagnosis. This undermines the generalization performance and the clinical applicability of the AI-based PC diagnosis approaches. Therefore, we propose a causal intervention based automated method for pancreatic cancer diagnosis with contrast-enhanced computerized tomography (CT) images, where a confounding effects reduction scheme is developed for alleviating spurious correlations to achieve unbiased learning, thereby improving the generalization performance. Specifically, a continuous image generation strategy was developed to simulate wide variations of intensity differences caused by imaging heterogeneities, where Monte Carlo sampling is added to further enhance the continuity of simulated images. Then, to enhance the pancreatic texture variability, a texture diversification method was introduced in conjunction with gradient-based data augmentation. Finally, a causal intervention strategy was proposed to alleviate the adverse confounding effects by decoupling the causal and non-causal factors and combining them randomly. Extensive experiments showed remarkable diagnosis performance on a cross-validation dataset. Also, promising generalization performance with an average accuracy of 0.87 was attained on three independent test sets of a total of 782 subjects. Therefore, the proposed method shows high clinical feasibility and applicability for pancreatic cancer diagnosis.

Real-time full-color meta fiber endoscopy.

The remarkable capacity of metasurfaces to exert precise control over the propagation of light has ushered in a realm of unprecedented possibilities for "Lab-on-fiber", and in this work, showcased real-time full-color imaging in a meta-optical fiber endoscope.

Exploring the microscopic changes of lipid droplets and mitochondria in alcoholic liver disease via fluorescent probes with high polarity specificity.

Alcoholic liver disease (ALD) has received extensive attention because of the increasing alcohol consumption globally as well as its high morbidity. It is reported that absorbed alcohol can cause lipid metabolism disorder and mitochondria dysfunction, so here in this work, we planned to study the microscopic changes of the two organelles, lipid droplets (LDs) and mitochondria in hepatocyte, under the stimulation of alcohol, hoping to present some meaningful information for the theranostics of ALD by the technique of fluorescence imaging. Guided by theoretical calculation, two fluorescent probes, named CBu and CBuT, were rationally designed. Although constructed by the same chromophore scaffold, they stained different organelles efficiently and emitted distinctively. CBu with high lipophilicity, ascribed to the two butyl groups, can selectively localize in LDs with green fluorescence, while CBuT bearing a triphenylphosphine unit can specifically target mitochondria due to electrostatic interactions with near-infrared (NIR) fluorescence. Both probes displayed remarkable selectivity and sensitivity to polarity, free from the environmental interferences including viscosity, pH and other bio-species. With these two probes, the accumulation of LDs and polarity decrease in mitochondria were clearly monitored at the green and red channels, respectively, in the ALD cell model. CBuT was further applied to image the mice with ALD in vivo. In short, we have confirmed the valuable organelles, LDs and mitochondria, for ALD study and provided two potent molecular tools to visualize their changes through fluorescence imaging, which would be favorable for the further development of theranostics for ALD.

Full-Stokes parameters detection enabled by a non-interleaved fiber-compatible metasurface.

Polarization of the optical field determines the way of light-matter interaction, which lays the foundation for various applications such as chiral spectroscopy, biomedical imaging, and machine vision. Currently, with the rise of the metasurface, miniaturized polarization detectors have attracted extensive interest. However, due to the limitation of the working area, it is still a challenge to integrate polarization detectors on the fiber end face. Here, we propose a design of compact non-interleaved metasurface that can be integrated on the tip of a large-mode-area photonic crystal fiber (LMA-PCF) to realize full-Stokes parameters detection. Through concurrent control over the dynamic phase and Pancharatnam-Berry (PB) phase, different helical phases are assigned to the two orthogonal circular polarization bases, of which the amplitude contrast and relative phase difference can be represented by two non-overlapped foci and an interference ring pattern, respectively. Therefore, the determination of arbitrary polarization states through the proposed ultracompact fiber-compatible metasurface can be achieved. Moreover, we calculated full-Stokes parameters according to simulation results and obtained that the average detection deviation is relatively low at 2.84% for 20 elucidated samples. The novel metasurface exhibits excellent polarization detection performance and overcomes the limitation of the small integrated area, which provides insights into the further practical explorations of ultracompact polarization detection devices.

Development and validation of an MRI-radiomics nomogram for the prognosis of pancreatic ductal adenocarcinoma.

Objective: To develop and validate an MRI-radiomics nomogram for the prognosis of pancreatic ductal adenocarcinoma (PDAC). Background: "Radiomics" enables the investigation of huge amounts of radiological features in parallel by extracting high-throughput imaging data. MRI provides better tissue contrast with no ionizing radiation for PDAC. Methods: There were 78 PDAC patients enrolled in this study. In total, there were 386 radiomics features extracted from MRI scan, which were screened by the least absolute shrinkage and selection operator algorithm to develop a risk score. Cox multivariate regression analysis was applied to develop the radiomics-based nomogram. The performance was assessed by discrimination and calibration. Results: The radiomics-based risk-score was significantly associated with PDAC overall survival (OS) (P < 0.05). With respect to survival prediction, integrating the risk score, clinical data and TNM information into the nomogram exhibited better performance than the TNM staging system, radiomics model and clinical model. In addition, the nomogram showed fine discrimination and calibration. Conclusions: The radiomics nomogram incorporating the radiomics data, clinical data and TNM information exhibited precise survival prediction for PDAC, which may help accelerate personalized precision treatment. Clinical trial registration: clinicaltrials.gov, identifier NCT05313854.

Generalized pancreatic cancer diagnosis via multiple instance learning and anatomically-guided shape normalization.

Pancreatic cancer is a highly malignant cancer type with a high mortality rate. As no obvious symptoms are associated with this cancer type, most of the diagnoses are made when the patients are already in a late stage. In this work, we propose an automated method for effective early diagnosis of pancreatic cancer based on multiple instance learning with contrast-enhanced CT images. In this method, diagnosis stability and generalizability were improved through shape normalization based on anatomical structures as well as instance-level contrastive learning. Specifically, anatomically-guided shape normalization were developed to reconstruct the pancreatic regions of interest by spatial transformations, account for larger tumor parts in these regions, and hence enhance the extraction of pancreatic features. Moreover, instance-level contrastive learning was employed to aggregate different types of tumor features within the multiple instance learning framework. This learning approach can maintain the tumor feature integrity and enhance the diagnosis stability. Finally, a balance-adjustment strategy was designed to alleviate the class imbalance problem caused by the scarcity of tumor samples. Extensive experimental results demonstrated remarkable performance of our method when conducted cross-validation on an in-house dataset with 310 patients and independent test on two unseen datasets (a private test set with 316 and a publicly-available test set with 281). The proposed strategies also led to significant improvements in generalizability. Besides, the clinical significance of the proposed method was further verified through two independent test results in which tumors smaller than 2 cm in diameter were identified at accuracies of 80.9% and 90.1%, respectively. Overall, our method provides a potentially successful tool for early diagnosis of pancreatic cancer. Our source codes will be released at https://github.com/SJTUBME-QianLab/MIL_PAdiagnosis.

Deep learning-based ballistocardiography reconstruction algorithm on the optical fiber sensor.

Ballistocardiography (BCG) is a vibration signal related to cardiac activity, which can be obtained in a non-invasive way by optical fiber sensors. In this paper, we propose a modified generative adversarial network (GAN) to reconstruct BCG signals by solving signal fading problems in a Mach-Zehnder interferometer (MZI). Based on this algorithm, additional modulators and demodulators are not needed in the MZI, which reduces the cost and hardware complexity. The correlation between reconstructed BCG and reference BCG is 0.952 in test data. To further test the model performance, we collect special BCG signals including sinus arrhythmia data and post-exercise cardiac activities data, and analyze the reconstructed results. In conclusion, a BCG reconstruction algorithm is presented to solve the signal fading problem in the optical fiber interferometer innovatively, which greatly simplifies the BCG monitoring system.

Optical Fiber-Integrated Metasurfaces: An Emerging Platform for Multiple Optical Applications.

The advent of metasurface technology has revolutionized the field of optics and photonics in recent years due to its capability of engineering optical wavefronts with well-patterned nanostructures at subwavelength scale. Meanwhile, inspired and benefited from the tremendous success of the "lab-on-fiber" concept, the integration of metasurface with optical fibers has drawn particular interest in the last decade, which establishes a novel technological platform towards the development of "all-in-fiber" metasurface-based devices. Thereby, this review aims to present and summarize the optical fiber-integrated metasurfaces with the current state of the art. The application scenarios of the optical fiber metasurface-based devices are well classified and discussed accordingly, with a brief explanation of physical fundamentals and design methods. The key fabrication methods corresponding to various optical fiber metasurfaces are summarized and compared. Furthermore, the challenges and potential future research directions of optical fiber metasurfaces are addressed to further leverage the flexibility and versatility of meta-fiber-based devices. It is believed that the optical fiber metasurfaces, as a novel all-around technological platform, will be exploited for a large range of applications in telecommunication, sensing, imaging, and biomedicine.

Dual-Wavelength Polarization-Dependent Bifocal Metalens for Achromatic Optical Imaging Based on Holographic Principle.

Recently, ultrathin metalenses have attracted dramatically growing interest in optical imaging systems due to the flexible control of light at the nanoscale. In this paper, we propose a dual-wavelength achromatic metalens that will generate one or two foci according to the polarization of the incident. Based on geometric phase modulation, two unit cells are attentively selected for efficient operation at distinct wavelengths. By patterning them to two divided sections of the metalens structure plane, the dual-wavelength achromatic focusing effect with the same focal length is realized. In addition, the holographic concept is adopted for polarization-dependent bifocal generation, in which the objective wave is originated from two foci that are respectively formed by two orthogonal polarization states of circularly polarized light, namely Left-handed circularly polarized (LCP) light and Right-handed circularly polarized (RCP) light. The incident light is considered as the reference light. The achromatic focusing and polarization-dependent bifocusing are numerically verified through simulations. The proposed design opens the path for the combination of multi-wavelength imaging and chiral imaging, which may find potential applications, such as achromatic optical devices and polarization-controlled biomedical molecular imaging systems.

Endless Single-Mode Photonics Crystal Fiber Metalens for Broadband and Efficient Focusing in Near-Infrared Range.

The advent of the 'lab-on-fiber' concept has boosted the prosperity of optical fiber-based platforms integrated with nanostructured metasurface technology which are capable of controlling the light at the nanoscale for multifunctional applications. Here, we propose an endless single-mode large-mode-area photonic crystal fiber (LMA-PCF) integrated metalens for broadband and efficient focusing from 800 to 1550 nm. In the present work, the optical properties of the substrate LMA-PCF were investigated, and the metalens, consisting of dielectric TiO2 nanorods with varying radii, was elaborately designed in the fiber core region with a diameter of 48 µm to cover the required phase profile for efficient focusing with a high transmission. The focusing characteristics of the designed metalens were also investigated in detail over a wide wavelength range. It is shown that the in-fiber metalens is capable of converging the incident beams into the bright, symmetric, and legible focal spots with a large focal length of 315-380 µm depending on the operating wavelength. A high and average focusing efficiency of 70% was also obtained with varying wavelengths. It is believed the proposed fiber metalens may show great potential in applications including fiber laser configuration, machining, and fiber communication.

Inhibition of DNM1L and mitochondrial fission attenuates inflammatory response in fibroblast-like synoviocytes of rheumatoid arthritis.

Mitochondrial fission and fusion are important for mitochondrial function, and dynamin 1-like protein (DNM1L) is a key regulator of mitochondrial fission. We investigated the effect of mitochondrial fission on mitochondrial function and inflammation in fibroblast-like synoviocytes (FLSs) during rheumatoid arthritis (RA). DNM1L expression was determined in synovial tissues (STs) from RA and non-RA patients. FLSs were isolated from STs and treated with a DNM1L inhibitor (mdivi-1, mitochondrial division inhibitor 1) or transfected with DNM1L-specific siRNA. Mitochondrial morphology, DNM1L expression, cell viability, mitochondrial membrane potential, reactive oxygen species (ROS), apoptosis, inflammatory cytokine expression and autophagy were examined. The impact of mdivi-1 treatment on development and severity of collagen-induced arthritis (CIA) was determined in mice. Up-regulated DNM1L expression was associated with reduced mitochondrial length in STs from patients with RA and increased RA severity. Inhibition of DNM1L in FLSs triggered mitochondrial depolarization, mitochondrial elongation, decreased cell viability, production of ROS, IL-8 and COX-2, and increased apoptosis. DNM1L deficiency inhibited IL-1ß-mediated AKT/IKK activation, NF-κBp65 nuclear translocation and LC3B-related autophagy, but enhanced NFKBIA expression. Treatment of CIA mice with mdivi-1 decreased disease severity by modulating inflammatory cytokine and ROS production. Our major results are that up-regulated DNM1L and mitochondrial fission promoted survival, LC3B-related autophagy and ROS production in FLSs, factors that lead to inflammation by regulating AKT/IKK/NFKBIA/NF-κB signalling. Thus, inhibition of DNM1L may be a new strategy for treatment of RA.

Circular Array of Magnetic Sensors for Current Measurement: Analysis for Error Caused by Position of Conductor.

This paper analyzes the measurement error, caused by the position of the current-carrying conductor, of a circular array of magnetic sensors for current measurement. The circular array of magnetic sensors is an effective approach for AC or DC non-contact measurement, as it is low-cost, light-weight, has a large linear range, wide bandwidth, and low noise. Especially, it has been claimed that such structure has excellent reduction ability for errors caused by the position of the current-carrying conductor, crosstalk current interference, shape of the conduction cross-section, and the Earth's magnetic field. However, the positions of the current-carrying conductor-including un-centeredness and un-perpendicularity-have not been analyzed in detail until now. In this paper, for the purpose of having minimum measurement error, a theoretical analysis has been proposed based on vector inner and exterior product. In the presented mathematical model of relative error, the un-center offset distance, the un-perpendicular angle, the radius of the circle, and the number of magnetic sensors are expressed in one equation. The comparison of the relative error caused by the position of the current-carrying conductor between four and eight sensors is conducted. Tunnel magnetoresistance (TMR) sensors are used in the experimental prototype to verify the mathematical model. The analysis results can be the reference to design the details of the circular array of magnetic sensors for current measurement in practical situations.

Non-obvious correlations to disease management unraveled by Bayesian artificial intelligence analyses of CMS data.

OBJECTIVE: Given the availability of extensive digitized healthcare data from medical records, claims and prescription information, it is now possible to use hypothesis-free, data-driven approaches to mine medical databases for novel insight. The goal of this analysis was to demonstrate the use of artificial intelligence based methods such as Bayesian networks to open up opportunities for creation of new knowledge in management of chronic conditions. MATERIALS AND METHODS: Hospital level Medicare claims data containing discharge numbers for most common diagnoses were analyzed in a hypothesis-free manner using Bayesian networks learning methodology. RESULTS: While many interactions identified between discharge rates of diagnoses using this data set are supported by current medical knowledge, a novel interaction linking asthma and renal failure was discovered. This interaction is non-obvious and had not been looked at by the research and clinical communities in epidemiological or clinical data. A plausible pharmacological explanation of this link is proposed together with a verification of the risk significance by conventional statistical analysis. CONCLUSION: Potential clinical and molecular pathways defining the relationship between commonly used asthma medications and renal disease are discussed. The study underscores the need for further epidemiological research to validate this novel hypothesis. Validation will lead to advancement in clinical treatment of asthma & bronchitis, thereby, improving patient outcomes and leading to long term cost savings. In summary, this study demonstrates that application of advanced artificial intelligence methods in healthcare has the potential to enhance the quality of care by discovering non-obvious, clinically relevant relationships and enabling timely care intervention.

Clinicopathologic features of patients with non-small cell lung cancer harboring the EML4-ALK fusion gene: a meta-analysis.

BACKGROUND: The frequencies of EML4-ALK fusion gene in non-small cell lung cancer (NSCLC) with different clinicopathologic features described by previous studies are inconsistent. The key demographic and pathologic features associated with EML4-ALK fusion gene have not been definitively established. This meta-analysis was conducted to compare the frequency of the EML4-ALK fusion gene in patients with different clinicopathologic features and to identify an enriched population of patients with NSCLC harboring EML4-ALK fusion gene. METHODS: The Pubmed and Embase databases for all studies on EML4-ALK fusion gene in NSCLC patients were searched up to July 2014. A criteria list and exclusion criteria were established to screen the studies. The frequency of the EML4-ALK fusion gene and the clinicopathologic features, including smoking status, pathologic type, gender, and EGFR status were abstracted. RESULTS: Seventeen articles consisting of 4511 NSCLC cases were included in this meta-analysis. A significant lower EML4-ALK fusion gene positive rate was associated with smokers (pooled OR = 0.40, 95% CI = 0.30-0.54, P<0.00001). A significantly higher EML4-ALK fusion gene positivity rate was associated with adenocarcinomas (pooled OR = 2.53, 95% CI = 1.66-3.86, P<0.0001) and female (pooled OR = 0.61, 95% CI = 0.41-0.90, P = 0.01). We found that a significantly lower EML4-ALK fusion gene positivity rate was associated with EGFR mutation (pooled OR = 0.07, 95% CI = 0.03-0.19, P<0.00001). No publication bias was observed in any meta-analysis (all P value of Egger's test >0.05); however, because of the small sample size, no results were in the meta-analysis regarding EGFR gene status. CONCLUSION: This meta-analysis revealed that the EML4-ALK fusion gene is highly correlated with a never/light smoking history, female and the pathologic type of adenocarcinoma, and is largely mutually exclusive of EGFR.

[A comparative study of Da Vinci robot system with video-assisted thoracoscopy in the surgical treatment of mediastinal lesions].

BACKGROUND: In recent years, Da Vinci robot system applied in the treatment of intrathoracic surgery mediastinal diseases become more mature. The aim of this study is to summarize the clinical data about mediastinal lesions of General Hospital of Shenyang Military Region in the past 4 years, then to analyze the treatment effect and promising applications of da Vinci robot system in the surgical treatment of mediastinal lesions. METHODS: 203 cases of mediastinal lesions were collected from General Hospital of Shenyang Military Region between 2010 and 2013. These patients were divided into two groups da Vinci and video-assisted thoracoscopic surgery (VATS) according to the selection of the treatments. The time in surgery, intraoperative blood loss, postoperative drainage amount within three days after surgery, the period of bearing drainage tubes, hospital stays and hospitalization expense were then compared. RESULTS: All patients were successfully operated, the postoperative recovery is good and there is no perioperative death. The different of the time in surgery between two groups is Robots group 82 (20-320) min and thoracoscopic group 89 (35-360) min (P>0.05). The intraoperative blood loss between two groups is robot group 10 (1-100) mL and thoracoscopic group 50 (3-1,500) mL. The postoperative drainage amount within three days after surgery between two groups is robot group 215 (0-2,220) mL and thoracoscopic group 350 (50-1,810) mL. The period of bearing drainage tubes after surgery between two groups is robot group 3 (0-10) d and thoracoscopic group: 5 (1-18) d. The difference of hospital stays between two groups is robot group 7 (2-15) d and thoracoscopic group 9 (2-50) d. The hospitalization expense between two groups is robot group (18,983.6±4,461.2) RMB and thoracoscopic group (9,351.9±2,076.3) RMB (All P<0.001). CONCLUSIONS: The da Vinci robot system is safe and efficient in the treatment of mediastinal lesions compared with video-assisted thoracoscopic approach, even though its expense is higher.

Primary malignant melanoma of the esophagus.

Primary malignant melanoma of the esophagus is an extremely rare, but highly aggressive, tumor. The prognosis for primary malignant melanoma of the esophagus remains dismal in most literatures. We report here a case of primary malignant melanoma of the esophagus that was treated by surgical resection and, additionally, followed by chemotherapy. The clinical features, treatment, pathological findings, and prognosis are analyzed and the literature of primary malignant melanoma of the esophagus is reviewed.

Long-term outcomes of video-assisted thoracic surgery lobectomy for nonsmall cell lung cancer.

OBJECTIVE: Although video-assisted thoracic surgery (VATS) lobectomy has been demonstrated to be safe and technically feasible for nonsmall cell lung cancer, it is infrequently performed in most hospitals. As a result, there has been no clear long-term outcome regarding this procedure until recently. METHODS: We retrospectively analyzed the results of patients suffering from lung cancer who underwent VATS lobectomy between March 1996 and December 2003. Perioperative data were collected and long-term outcomes were assessed by a 5-year census. RESULTS: We successfully performed VATS procedures in 92 of 97 patients. The median operation time was 168 minutes and the median hospitalization period was 8.5 days. The median operative blood loss was 110 mL and the median drainage time was 3.3 days. The operative and postoperative (30-day) mortalities were 0% and 1.1%, respectively. The postoperative complications rate was 10.9%. The 5-year overall survival rates for stages I, II, and III or greater were 81.7%, 43.5%, and 22.2%, respectively. CONCLUSIONS: VATS lobectomy is a safe and feasible surgical procedure for nonsmall cell lung cancer and gives the same long-term outcomes as conventional open lobectomy.

Video-assisted thoracic surgery (VATS) compares favorably with thoracotomy for the treatment of lung cancer: a five-year outcome comparison.

BACKGROUND: Although video-assisted thoracic surgery (VATS) lobectomy has been demonstrated to be safe and technically feasible, it is infrequently performed in most Chinese hospitals and few thoracic surgeons have performed the operation. We have therefore reviewed our experience with all VATS lobectomies, attempting to define long- and short-term outcomes of these surgeries. METHODS: We retrospectively analyzed the results in patients who underwent VATS lobectomy and open lobectomy between March 1996 and August 2003. The VATS surgery was performed with the endoscopic hilar dissection technique. Perioperative data were collected and long-term outcomes were assessed by 5-year census. RESULTS: We successfully performed VATS procedures in 113 of 120 patients. The median operative time was 152 min and the median hospitalization was 8.6 days. The median operative blood loss was 130 ml and the median drainage time was 3.5 days. The operative and perioperative (30-day) mortality rates were 0% and 0.9%, respectively; the postoperative complications rate was 10.6%. The 5-year overall survival rates for stage I, stage II, and stage III or greater non-small cell lung cancer (NSCLC), secondary pulmonary malignancy, and benign disease were 79.1%, 45.5%, 22.2%, 33.3%, and 88.6%, respectively. CONCLUSIONS: Video-assisted lobectomy is a safe and feasible surgical procedure, and it gives the same long-term results as conventional open lobectomy. The VATS approach to lobectomy is a beneficial alternative to standard thoracotomy for selected cases of pulmonary lesions.

[Video-assisted thoracic surgery: clinical experience among 1264 patients].

OBJECTIVE: To summarize the clinical experience in video-assisted thoracic surgery (VATS). METHODS: From December 1993 to December 2005 1264 patients, 894 males and 370 females, aged 38.9 +/- 12.0, underwent VATS, including bullectomy in 622 cases, resection of mediastinal tumor or cyst in 119 cases, resection of esophageal diseases in 107 cases, lobectomy or wedge-shaped lung resection in 215 cases, lung volume reduction surgery (LVRS) in 17 cases, treatment of thoracic injury in 28 cases, treatment of other thoracic diseases in 72 cases, and biopsy in 84 cases. For the resection of esophageal carcinoma VATS was conducted via the right approach, the esophagus was dissociated, the lymph nodes were resected, upper-abdominal incision was made, the stomach was dissociated and drawn up to the neck region, a cervical incision was made to anastomose the stomach and the residue of esophagus. RESULTS: Operation was completed by VATS successfully in 1230 patients, and 34 cases were converted to traditional thoracotomy because of thoracic adhesion or to radically treat the malignant tumors. Major complications occurred in 45 cases (3.56%), including air-leak lasting more than 7 days in 30 cases, post-operative bleeding in 4 cases (3 of which received VATS once more for hemostasis and the other underwent thoracotomy), hydrothorax or pneumothorax in 3 cases that underwent water-closed drainage, esophageal mucous rupture in 4 cases with achalasia and one case with leiomyoma, all of which underwent repair immediately, infection of pleural cavity in one case after the resection of esophageal diverticulum, and pneumonia in one case after LVRS. One patient with spontaneous pneumothorax and respiratory failure died 5 days after the bullectomy. Spontaneous pneumothorax occurred in 10 patients 2 months to 2 years after VATS 3 of which underwent bullectomy and pleurodesis by VATS once more. CONCLUSION: Spontaneous pneumothorax and some benign thoracic diseases are the major indications of VATS; however, great care should be expended to decide to treat malignant diseases by VATS. It is very important to train the surgeons who are to practice VATS. The practice of VATS should be individualized.