Native-lung complications following single-lung transplantation for interstitial lung disease: an in-depth analysis.

BACKGROUND: Interstitial lung disease (ILD) represents a heterogeneous group of lung disorders characterized by fibrotic lung tissue changes. In regions with severe donor shortages, single-lung transplantation (SLTx) is often preferred over bilateral lung transplantation for advanced ILD. However, temporal changes and complications in the retained native lung remain poorly understood. METHODS: A retrospective analysis of 149 recipients who had undergone SLTx was conducted, including 34 ILD SLTx recipients. Native-lung volume, radiological alterations, and perfusion were assessed at distinct post-SLTx time points. Statistical analyses compared ILD and non-ILD SLTx groups. RESULTS: Our study revealed a progressive reduction in native-lung volume over time, accompanied by radiographic deterioration and declining perfusion. Complications in the retained native lung were observed, such as pneumothorax (29.4%), pulmonary aspergillosis (11.8%), and acute exacerbation (8.9%). Long-term survival rates were similar between ILD and non-ILD SLTx recipients. CONCLUSIONS: This study illuminates the unique challenges and complications with respect to the native lung following SLTx for ILD. Ongoing monitoring and tailored management are essential. Despite limitations, this research contributes to our understanding of the temporal progression of native-lung complications post-SLTx for ILD, underscoring the need for further investigation.

Real-time ultrasound-guided thoracentesis simulation using an optical see-through head-mounted display: a proof-of-concept study.

Aim: This study aimed to examine the feasibility and potential benefits of an optical see-through head-mounted display (OST-HMD) during real-time ultrasound-guided thoracentesis simulations. Material and methods: Six physicians performed a thoracentesis simulation using an OST-HMD and a wireless image transmission system. The time required, puncture needle visibility, pleural fluid collection success rate, and head movement during the procedure using a smart glass equipped with an inertial measurement unit were all recorded and compared with and without the HMD. Results: Study participants successfully extracted effusions in all procedures. The use of OST-HMD did not significantly affect the time of the procedure, but notably decreased the horizontal and vertical head movements during the procedure. Conclusions: The study demonstrated the feasibility of using an OST-HMD in a simulated real-time ultrasound-guided thoracentesis procedure and showed the potential of HMD in thoracentesis to improve ergonomics and accuracy. Further research is necessary to confirm these findings.

Autopsy case: Pathological complete response in an advanced lung cancer patient with severe immune related adverse events and COVID-19.

A 51-year-old man with advanced lung cancer underwent chemotherapy including an immune checkpoint inhibitor (ICI). Subsequently, he developed ICI-related pneumonitis and colitis, followed by a ten-month treatment course of high doses of steroids. An infection with the SARS-CoV-2 Omicron BA.5 variant caused lethal respiratory failure. Autopsy examination revealed no evidence of viable lung cancer cells in the evaluated organs including the lungs, suggesting that pathological complete response (pCR) could be successfully achieved by ICI treatment in advanced lung cancer patients. Recognizing steroid-resistant ICI-related adverse events as critical factors in severe COVID-19 emphasizes the need for appropriate assessment of ICI-induced pCR.

Effect of Liver Dysfunction on S-1 Therapy Induced Adverse Effects: A Retrospective Cohort Study.

BACKGROUND/AIM: Renal dysfunction necessitates S-1 dose reduction. However, decreased dihydropyrimidine dehydrogenase (DPD) activity may lead to adverse events due to 5-FU. The guidelines provided by pharmaceutical companies state that total bilirubin (T-Bil) should be ≤upper limit of normal (ULN)×1.5 as a reference value for safely taking S-1. Nevertheless, the relationship between the degree of liver dysfunction and S-1 dose reduction has not been clearly established. PATIENTS AND METHODS: This study focused on patients who received S-1 monotherapy for various types of cancer. The primary outcome was defined as the variation between blood sampling results on the test day and the subsequent test. The variation data were categorized based on the difference in T-Bil: Low T-Bil group (≤2.25) and High T-Bil group (>2.25). RESULTS: The number of patients that underwent S-1 monotherapy was 883 and the running number was 7,511; Low T-Bil group included 7,245 and High T-Bil group included 266. Examination of the effect of the T-Bil Group on clinical outcomes revealed a correlation with red blood cell (RBC) count, platelet (PLT) count, and T-Bil level. When the impact of the interaction between the T-Bil Group and any of the clinical outcomes, such as the RBC count, PLT count, and T-Bil level, was determined, each outcome showed a significant decrease in the High T-Bil group compared with the Low T-Bil group. CONCLUSION: S-1 administration in patients with liver dysfunction accompanied by elevated T-Bil levels may cause thrombocytopenia.

Optical Biopsies Using Probe-Based Confocal Laser Endomicroscopy for Autoimmune Pulmonary Alveolar Proteinosis.

INTRODUCTION: Increasing numbers of cases of mild asymptomatic pulmonary alveolar proteinosis (PAP) are being reported with the recent increase in chest computed tomography (CT). Bronchoscopic diagnosis of mild PAP is challenging because of the patchy distribution of lesions, which makes it difficult to obtain sufficient biopsy samples. Additionally, the pathological findings of mild PAP, particularly those that differ from severe PAP, have not been fully elucidated. This study aimed to clarify the pathological findings of mild PAP and the usefulness of optical biopsy using probe-based confocal laser endomicroscopy (pCLE). METHODS: We performed bronchoscopic optical biopsy using pCLE and tissue biopsy in 5 consecutive patients with PAP (three with mild PAP and two with severe PAP). We compared the pCLE images of mild PAP with those of severe PAP by integrating clinical findings, tissue pathology, and chest CT images. RESULTS: pCLE images of PAP showed giant cells with strong fluorescence, amorphous substances, and thin alveolar walls. Images of affected lesions in mild PAP were equivalent to those obtained in arbitrary lung lesions in severe cases. All 3 patients with mild PAP spontaneously improved or remained stable after ≥3 years of follow-up. Serum autoantibodies to granulocyte-macrophage colony-stimulating factor were detected in all 5 cases. CONCLUSION: Optical biopsy using pCLE can yield specific diagnostic findings, even in patients with mild PAP. pCLE images of affected areas in mild and severe PAP showed similar findings, indicating that the dysfunction level of pathogenic alveolar macrophages in affected areas is similar between both disease intensities.

Effectiveness of remdesivir-based therapy for moderate COVID-19: comparison of Omicron and other variant phases.

Remdesivir is an antiviral drug for the treatment of coronavirus disease 2019 (COVID-19), and the sustained antiviral activity against Omicron variants of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has been reported. In this single-center retrospective study, we first compared the clinical effectiveness of remdesivir-based therapy between Omicron and other variant phases of moderate COVID-19 in a real-world setting. Between Dec 2020 and July 2022, a total of 406 patients with COVID-19 pneumonia were treated with remdesivir-based therapy on admission. The oxygen deterioration rate after initiation of treatment significantly decreased in the Omicron variant phase compared to the alpha and delta variant phases. In an adjusted multivariate Cox proportional hazards model, Omicron variant phase was significantly associated with delayed oxygen deterioration and early recovery from hypoxia. These favorable outcomes during the Omicron variant phase, compared to previous variant phases, might be due to the attenuation and the popularization of vaccination.

Efficacy and safety of second-line therapy of docetaxel plus ramucirumab after first-line platinum-based chemotherapy plus immune checkpoint inhibitors in non-small cell lung cancer (SCORPION): a multicenter, open-label, single-arm, phase 2 trial.

Background: Immune checkpoint inhibitors (ICI) plus platinum-based chemotherapy has been recognized as a standard first-line therapy in non-small cell lung cancer (NSCLC); however, no prospective clinical trials of docetaxel (DTX) plus ramucirumab (RAM) following first-line ICI plus platinum-based chemotherapy has been reported. Methods: In this multicentre, open-label, single-arm, phase 2 trial, we enrolled patients with NSCLC from eight centres in Japan. Patients with metastatic NSCLC with disease progression after platinum-based chemotherapy plus ICI were eligible for the study. Patients were intravenously treated with 60 mg/m2 of DTX and 10 mg/kg of RAM on day 1 with a strong recommendation of pegfilgrastim administration on day 2 every 3 weeks. The primary end point was objective response rate (ORR) in efficacy analysis population. Safety was assessed in all patients treated at least one dose. The ORR of the null and alternative hypotheses were 10% and 30%, with α error of 0.1 and ß error of 0.1. This trial is registered with the Japan Registry for Clinical Trials, jCRTs041190077. Findings: Between 16 January, 2020, and 24 August, 2021, 33 patients (median age 66 [range 42-79] years) were enrolled. Thirteen patients (41%) had Eastern Cooperative Oncology Group performance status of 1. Twenty-five patients (78%) had an interval of <60 days after the last administration of ICI. In the efficacy analysis population (n = 32), the primary endpoint was met as 11 patients achieved partial response (PR), with ORR of 34.4% (80% CI, 23.1-47.2). Grade ≥3 anaemia and febrile neutropenia were observed in 2 (6%) and 3 (9%) patients, respectively. No treatment-related deaths and no new safety signals were observed. Interpretation: DTX plus RAM demonstrated encouraging antitumor activity with a manageable safety profile in patients who have progressed on front-line ICIs plus platinum-based chemotherapy. The results of this trial can be a helpful reference in conducting further phase III trials of new second-line treatment options. Funding: Eli Lilly Japan K.K.

Acute exacerbation of rheumatoid arthritis-associated interstitial lung disease triggered by COVID-19: What is the best practice for treatment?

We present a case of 79-year-old female with rheumatoid arthritis-associated interstitial lung disease (RA-ILD) developed an acute exacerbation (AE) triggered by coronavirus disease 2019 (COVID-19). The patient was unresponsive to a combination therapy of remdesivir, dexamethasone, and tocilizumab. Given that a recent multicenter cohort study reported ILD as a poor prognostic contributor in patients with RA and COVID-19, there may be potentially a certain number of patients with AE of RA-ILD triggered by COVID-19. This case highlights the need for a discussion how to treat these patients in a daily clinical practice.

A Retrospective, Multicenter, Observational Study to Evaluate Clinical Outcomes of Lorlatinib After Alectinib in Patients With ALK-Positive NSCLC in Japan.

Introduction: Lorlatinib is an ALK tyrosine kinase inhibitor approved in Japan for the treatment of advanced ALK+ NSCLC. There has been little evidence about lorlatinib efficacy after first-line (1L) alectinib in clinical practice in Japan. Methods: We retrospectively analyzed patients with advanced ALK+ NSCLC previously treated with 1L alectinib at multiple sites in Japan. Primary objectives were to collect patient demographics at baseline and estimate time to treatment failure (TTF) with second-line (2L) or third-line (3L) or later line (≥3L) lorlatinib treatment. Secondary objectives included objective response rate (ORR) with lorlatinib, reason for discontinuation and time to last treatment failure with lorlatinib, TTF and ORR of alectinib, and combined TTF. Results: Among the 51 patients included in the study, 29 (56.9%) received 2L and 22 (43.1%) received ≥3L lorlatinib treatment. At lorlatinib initiation, brain metastases were reported in 25 patients (49.0%), and 32 (62.7%) had an Eastern Cooperative Oncology Group performance status of 0 or 1. Median TTF with lorlatinib was 11.1 months (95% confidence interval [CI]: 4.6-13.8) in any line, 10.8 months (95% CI: 3.9-13.8) in 2L, and 11.5 months (95% CI: 2.9-not reached) in ≥3L. Median TTF was 11.5 months (95% CI: 3.9-not reached) in patients with brain metastases at lorlatinib initiation and 9.9 months (95% CI: 4.3-13.8) in patients without brain metastases. ORR was 35.7% with any-line lorlatinib treatment. Conclusions: Patient characteristics and efficacy were comparable with previous reports when lorlatinib was given after 1L alectinib in patients with ALK+ NSCLC.

Factors Affecting the Comprehension of Outpatients Receiving Cancer Chemotherapy.

Pharmaceutical consultation targeting outpatients at the Fujita Health University Hospital (Japan) provides support to patients undergoing anticancer drug treatment. This study aimed to explore factors that affect the comprehension of cancer chemotherapy among outpatients who received cancer treatment at our hospital. A questionnaire survey was conducted, and comprehension was scored on a scale of 1-5 (1, no comprehension; 5, full comprehension). When factors other than age and sex [the influence of which on comprehension has been reported in previous reports] were noted, differences in comprehension between the questionnaire items were comparatively analyzed according to the presence/absence of the relevant factors. Overall, 536 patients were included. Age (<70 years) and pharmacist interventions were identified as factors contributing to a comprehension score. The levels of comprehension regarding the name of the cancer chemotherapy, content/schedule of the treatment, purposes of the prescribed drugs, and objectives of blood tests were significantly higher in the group that received the pharmaceutical interventions; conversely, the level of comprehension for the self-management of adverse events was significantly lower in this group than in the group that did not receive any pharmaceutical interventions. Age and interventions by the pharmacist affected the comprehension of cancer chemotherapy by patients.

COVID-19 symptom-onset to diagnosis and diagnosis to treatment intervals are significant predictors of disease progression and hospitalization in high-risk patients: A real world analysis.

BACKGROUND: Coronavirus disease (COVID-19) is overwhelming healthcare systems worldwide. This study aimed to elucidate factors that influence disease progression to pneumonia and hospitalization before and after antiviral treatment for COVID-19 in an outpatient setting. METHODS: A total of 206 high-risk patients with COVID-19 were treated with sotrovimab, remdesivir, and molnupiravir at the Toshiwakai clinic between January 4 and April 30, 2022. Of these, 49 patients visited the Toshiwakai clinic directly and were treated immediately after diagnosis (Toshiwakai-clinic study group). The remaining patients were diagnosed elsewhere, and of these, 102 patients were quarantined at home (health-center study group) and 55 at designated facilities (quarantine-facility study group) before being referred to Toshiwakai clinic. Patients were categorized into those with mild and moderate COVID-19, based on the presence of pneumonia at the initial visit to Toshiwakai clinic. RESULTS: The symptom-onset-to-diagnosis and diagnosis-to-treatment intervals were significant predictors of moderate disease. Age, dyspnea, and diagnosis-to-treatment interval at the first visit to Toshiwakai clinic were significant predictors for hospitalization even after antiviral treatment. Although the symptom-onset-to-diagnosis interval did not differ among the three study groups, the diagnosis-to-treatment and symptom-onset-to-treatment intervals were significantly longer in the health-center and quarantine-facility study groups than in the Toshiwakai-clinic study group. CONCLUSION: The symptom-onset-to-diagnosis and diagnosis-to-treatment intervals reflect diagnostic and interventional delays, respectively, which are closely related to the current COVID-19 clinical management protocol. Easy access to the clinics and immediate antiviral treatment after diagnosis may be the best methods to prevent disease progression and hospitalization in high-risk patients.

Clinical characteristics and courses of 200 patients hospitalized for COVID-19 during the second and third waves at Fujita Health University Okazaki Medical Center in Japan.

Objectives: There are few reports about patients hospitalized for COVID-19 in Japan. We investigated 200 patients hospitalized for COVID-19 over a 6-month period with the aim of elucidating their clinical characteristics and clinical courses. Methods: The study cohort comprised 200 patients hospitalized for COVID-19 during a 6-month period. We examined baseline characteristics, source of transmission, preadmission course, initial symptoms, concomitant symptoms, comorbidities, treatments, and prognosis. Results: The number of inpatients from outside our region increased from 9 in the second wave to 53 in the third wave. The initial manifestations were cold-like and gastroenteritis-like symptoms, gustatory and olfactory dysfunction being frequently occurring concomitant symptoms. On admission 32 patients had mild disease, 108 moderate I, 54 moderate II, and 6 severe. We divided the 200 patients into second and third wave groups and compared their baseline characteristics. The third wave group was older and had more severe disease. The main treatments implemented were dexamethasone and remdesivir. Three patients (1.5%) required ventilation and 12 (6.0%) died in hospital. Conclusions: We investigated 200 patients hospitalized for COVID-19 over a period of 6 months. The patients in the second wave were relatively young and most had mild disease. In contrast, the patients in the third wave were older and had more severe disease and higher in-hospital mortality.

Post-discharge clinical, laboratory and radiographic features of coronavirus disease 2019 (COVID-19) patients at university hospitals in Japan.

Objectives: Coronavirus disease 2019 (COVID-19) has affected nearly half million people in Japan. However, information on the prolonged symptoms as well as laboratory and radiographic findings after hospital discharge remains limited. Methods: We retrospectively collected the symptoms, laboratory test results, and chest imaging results of COVID-19 patients at the time of the hospital admission and the ambulatory visits after discharge at two university hospitals between July and December 2020. Patients: A total of 126 COVID-19 patients, including of 88 with mild to moderate disease and 38 with severe to critical disease, were included. The time between symptom onset and the first outpatient visit was 46 days (Interquartile range, 39 to 55). Results: At the ambulatory visits, 36.5% of patients had at least one symptom. The most frequent symptom was shortness of breath (12.8%), followed by cough (11.1%), and fatigue (8.8%). Of 120 patients with post-discharge laboratory test results, 27 patients (22.5%) had abnormal alanine aminotransferase levels, and 35 patients (29.1%) had lymphocytopenia, including 24 and 27 mild and moderate patients. Of 122 patients with post-discharge chest computed tomography (CT) scans, 105 (83.3%) had abnormal findings. This abnormality was found in both mild to moderate and severe patients. Conclusions: Shortness of breath, abnormal liver function test results and chest CT images often persisted for at least one month after discharge, even when symptoms were mild or moderate during hospitalization.

Automated Classification of Idiopathic Pulmonary Fibrosis in Pathological Images Using Convolutional Neural Network and Generative Adversarial Networks.

Interstitial pneumonia of uncertain cause is referred to as idiopathic interstitial pneumonia (IIP). Among the various types of IIPs, the prognosis of cases of idiopathic pulmonary fibrosis (IPF) is extremely poor, and accurate differentiation between IPF and non-IPF pneumonia is critical. In this study, we consider deep learning (DL) methods owing to their excellent image classification capabilities. Although DL models require large quantities of training data, collecting a large number of pathological specimens is difficult for rare diseases. In this study, we propose an end-to-end scheme to automatically classify IIPs using a convolutional neural network (CNN) model. To compensate for the lack of data on rare diseases, we introduce a two-step training method to generate pathological images of IIPs using a generative adversarial network (GAN). Tissue specimens from 24 patients with IIPs were scanned using a whole slide scanner, and the resulting images were divided into patch images with a size of 224 × 224 pixels. A progressive growth GAN (PGGAN) model was trained using 23,142 IPF images and 7817 non-IPF images to generate 10,000 images for each of the two categories. The images generated by the PGGAN were used along with real images to train the CNN model. An evaluation of the images generated by the PGGAN showed that cells and their locations were well-expressed. We also obtained the best classification performance with a detection sensitivity of 97.2% and a specificity of 69.4% for IPF using DenseNet. The classification performance was also improved by using PGGAN-generated images. These results indicate that the proposed method may be considered effective for the diagnosis of IPF.

Syphilis With Multiple Pulmonary Nodules and Bone Lesions Detected on Computed Tomography and 18F Fluorodeoxyglucose-Positron Emission Tomography/Computed Tomography.

Syphilis can cause a wide range of systemic manifestations, such as papular rash, malaise, weight loss, muscle aches, generalized lymphadenopathy, and meningitis. However, pulmonary involvement in patients with secondary syphilis is thought to be relatively rare. Moreover, bone involvement in patients with secondary syphilis is also considered rare, and only a few cases of involvement of lung and bone in such patients have been reported. In this paper, we report a case of secondary syphilis with pulmonary involvement in the form of multiple nodules with low attenuation areas, lymphadenopathy and multiple bone lesions detected on computed tomography and 18F fluorodeoxyglucose-positron emission tomography/computed tomography.

Iris-derived induced pluripotent stem cells that express GFP in all somatic cells of mice and differentiate into functional retinal neurons.

When regenerated tissue is generated from induced pluripotent stem cells (iPSCs), it is necessary to track and identify the transplanted cells. Fluorescently-labeled iPSCs synthesize a fluorescent substance that is easily tracked. However, the expressed protein should not affect the original genome sequence or pluripotency. To solve this problem, we created a cell tool for basic research on iPSCs. Iris tissue-derived cells from GFP fluorescence-expressing mice (GFP-DBA/2 mice) were reprogrammed to generate GFP mouse iris-derived iPSCs (M-iris GFP iPSCs). M-iris GFP iPSCs expressed cell markers characteristic of iPSCs and showed pluripotency in differentiating into the three germ layers. In addition, when expressing GFP, the cells differentiated into functional recoverin- and calbindin-positive cells. Thus, this cell line will facilitate future studies on iPSCs.

Adverse reactions to mRNA coronavirus disease 2019 (COVID-19) vaccine for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in 576 medical staff.

Objectives: Various measures have been taken to curb the COVID-19 epidemic, but their effect has been limited. Vaccines are expected to make a definite improvement. In Japan, vaccination began in February 2021. However, there are very few reports of adverse reactions to mRNA vaccines for the new coronavirus (SARS-CoV-2) in Japanese people. Therefore, adverse reactions in 576 vaccinated medical staff at the author's hospital were investigated. Methods: The subjects were 576 medical staff who wished to receive the new coronavirus (SARS-CoV-2) vaccine. The first vaccination was performed with Pfizer's new coronavirus mRNA vaccine (commercial name: Comirnaty intramuscular injection) from March 8 to March 15, 2021. The second vaccination was performed from March 29 to April 5, 2021, at an interval of more than 21 days from the first vaccination. Results: Adverse reactions were seen in six subjects in the first vaccination. There was dyspnea in one subject, arthralgia in one subject, fever/malaise in three subjects, and left axillary pain and lymphadenopathy in one subject. One subject had a reaction at the vaccination site that needed to be addressed. At the second vaccination, adverse reactions were observed in 64 subjects. There was fever in 58 subjects, malaise in 21 subjects, arthralgia in 12 subjects, aching pain in 11 subjects, headache in six subjects, chills in six subjects, nausea in three subjects, redness in two subjects, dizziness in two subjects, hives in two subjects, swelling in two subjects, cough in one subject, and itching in one subject (some had multiple adverse reactions). Fever was observed in the range of body temperature from 37.4 to 38.9 degrees. Conclusions: Most of the adverse reactions to the COVID-19 vaccine were mild, and no serious anaphylaxis was observed. Vaccination was considered perfectly feasible if attention is paid to adverse reactions.

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.

Formation of three-dimensional cell aggregates expressing lens-specific proteins in various cultures of human iris-derived tissue cells and iPS cells.

Induced pluripotent stem (iPS) cells are widely used as a research tool in regenerative medicine and embryology. In studies related to lens regeneration in the eye, iPS cells have been reported to differentiate into lens epithelial cells (LECs); however, to the best of our knowledge, no study to date has described their formation of three-dimensional cell aggregates. Notably, in vivo studies in newts have revealed that iris cells in the eye can dedifferentiate into LECs and regenerate a new lens. Thus, as basic research on lens regeneration, the present study investigated the differentiation of human iris tissue-derived cells and human iris tissue-derived iPS cells into LECs and their formation of three-dimensional cell aggregates using a combination of two-dimensional culture, static suspension culture and rotational suspension culture. The results revealed that three-dimensional cell aggregates were formed and differentiated into LECs expressing αA-crystallin, a specific marker protein for LECs, suggesting that the cell-cell interaction facilitated by cell aggregation may have a critical role in enabling highly efficient differentiation of LECs. However, the present study was unable to achieve transparency in the cell aggregates; therefore, we aim to continue to investigate the degradation of organelles and other materials necessary to make the interior of the formed cell aggregates transparent. Furthermore, we aim to expand on our current work to study the regeneration of the lens and ciliary body as a whole in vitro, with the aim of being able to restore focusing function after cataract surgery.

[Prognosis Prediction of Lung Cancer Patients Using CT Images: Feature Extraction by Convolutional Neural Network and Prediction by Machine Learning].

PURPOSE: Lung cancer accounts for the largest number of deaths among malignant tumors. Recently, more and more patients are concerned about their own life expectancy. CT examination is essential for the diagnosis of lung cancer. However, it is difficult to accurately predict the prognosis using CT images. In this study, we developed a method to predict the prognosis of lung cancer patients from CT images using a convolutional neural network (CNN) and a machine learning method. METHODS: In this study, the CT images of 173 lung cancer patients were collected. First, we selected the slice with the largest tumor size in each case and extracted features using a CNN. Next, we performed feature selection using information gain and predicted alive or death by classifiers. An artificial neural network or Naïve Bayes was used as a classifier and alive and death were predicted at one-year intervals from one year to five years later. RESULTS: We evaluated the prediction accuracy via the three-fold cross-validation method and found that the prediction accuracies were around 80% for all periods from 1 to 5 years. In the evaluation of the survival curve, the shape of the curve was close to the actual curve. CONCLUSION: These results indicate that feature extraction by a CNN and classification by the machine learning method may be effective in predicting the prognosis of lung cancer patients using CT images.