Standalone deep learning versus experts for diagnosis lung cancer on chest computed tomography: a systematic review.

PURPOSE: To compare the diagnostic performance of standalone deep learning (DL) algorithms and human experts in lung cancer detection on chest computed tomography (CT) scans. MATERIALS AND METHODS: This study searched for studies on PubMed, Embase, and Web of Science from their inception until November 2023. We focused on adult lung cancer patients and compared the efficacy of DL algorithms and expert radiologists in disease diagnosis on CT scans. Quality assessment was performed using QUADAS-2, QUADAS-C, and CLAIM. Bivariate random-effects and subgroup analyses were performed for tasks (malignancy classification vs invasiveness classification), imaging modalities (CT vs low-dose CT [LDCT] vs high-resolution CT), study region, software used, and publication year. RESULTS: We included 20 studies on various aspects of lung cancer diagnosis on CT scans. Quantitatively, DL algorithms exhibited superior sensitivity (82%) and specificity (75%) compared to human experts (sensitivity 81%, specificity 69%). However, the difference in specificity was statistically significant, whereas the difference in sensitivity was not statistically significant. The DL algorithms' performance varied across different imaging modalities and tasks, demonstrating the need for tailored optimization of DL algorithms. Notably, DL algorithms matched experts in sensitivity on standard CT, surpassing them in specificity, but showed higher sensitivity with lower specificity on LDCT scans. CONCLUSION: DL algorithms demonstrated improved accuracy over human readers in malignancy and invasiveness classification on CT scans. However, their performance varies by imaging modality, underlining the importance of continued research to fully assess DL algorithms' diagnostic effectiveness in lung cancer. CLINICAL RELEVANCE STATEMENT: DL algorithms have the potential to refine lung cancer diagnosis on CT, matching human sensitivity and surpassing in specificity. These findings call for further DL optimization across imaging modalities, aiming to advance clinical diagnostics and patient outcomes. KEY POINTS: Lung cancer diagnosis by CT is challenging and can be improved with AI integration. DL shows higher accuracy in lung cancer detection on CT than human experts. Enhanced DL accuracy could lead to improved lung cancer diagnosis and outcomes.

Progress in Serial Imaging for Prognostic Stratification of Lung Cancer Patients Receiving Immunotherapy: A Systematic Review and Meta-Analysis.

Immunotherapy, particularly with checkpoint inhibitors, has revolutionized non-small cell lung cancer treatment. Enhancing the selection of potential responders is crucial, and researchers are exploring predictive biomarkers. Delta radiomics, a derivative of radiomics, holds promise in this regard. For this study, a meta-analysis was conducted that adhered to PRISMA guidelines, searching PubMed, Embase, Web of Science, and the Cochrane Library for studies on the use of delta radiomics in stratifying lung cancer patients receiving immunotherapy. Out of 223 initially collected studies, 10 were included for qualitative synthesis. Stratifying patients using radiomic models, the pooled analysis reveals a predictive power with an area under the curve of 0.81 (95% CI 0.76-0.86, p < 0.001) for 6-month response, a pooled hazard ratio of 4.77 (95% CI 2.70-8.43, p < 0.001) for progression-free survival, and 2.15 (95% CI 1.73-2.66, p < 0.001) for overall survival at 6 months. Radiomics emerges as a potential prognostic predictor for lung cancer, but further research is needed to compare traditional radiomics and deep-learning radiomics.

Evaluating the Potential of Delta Radiomics for Assessing Tyrosine Kinase Inhibitor Treatment Response in Non-Small Cell Lung Cancer Patients.

Our study aimed to harness the power of CT scans, observed over time, in predicting how lung adenocarcinoma patients might respond to a treatment known as EGFR-TKI. Analyzing scans from 322 advanced stage lung cancer patients, we identified distinct image-based patterns. By integrating these patterns with comprehensive clinical information, such as gene mutations and treatment regimens, our predictive capabilities were significantly enhanced. Interestingly, the precision of these predictions, particularly related to radiomics features, diminished when data from various centers were combined, suggesting that the approach requires standardization across facilities. This novel method offers a potential pathway to anticipate disease progression in lung adenocarcinoma patients treated with EGFR-TKI, laying the groundwork for more personalized treatments. To further validate this approach, extensive studies involving a larger cohort are pivotal.

Undiagnosed dyspnea with pulmonary nodule.

Physicians should remain vigilant about alternative causes of shortness of breath even when respiratory diseases are being effectively treated. The lateral view of chest radiography can be valuable in discerning the three-dimensional characteristics of pulmonary abnormalities.

Application of Radiomics in Prognosing Lung Cancer Treated with Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitors: A Systematic Review and Meta-Analysis.

In the context of non-small cell lung cancer (NSCLC) patients treated with EGFR tyrosine kinase inhibitors (TKIs), this research evaluated the prognostic value of CT-based radiomics. A comprehensive systematic review and meta-analysis of studies up to April 2023, which included 3111 patients, was conducted. We utilized the Quality in Prognosis Studies (QUIPS) tool and radiomics quality scoring (RQS) system to assess the quality of the included studies. Our analysis revealed a pooled hazard ratio for progression-free survival of 2.80 (95% confidence interval: 1.87-4.19), suggesting that patients with certain radiomics features had a significantly higher risk of disease progression. Additionally, we calculated the pooled Harrell's concordance index and area under the curve (AUC) values of 0.71 and 0.73, respectively, indicating good predictive performance of radiomics. Despite these promising results, further studies with consistent and robust protocols are needed to confirm the prognostic role of radiomics in NSCLC.

A radiomics-based deep learning approach to predict progression free-survival after tyrosine kinase inhibitor therapy in non-small cell lung cancer.

BACKGROUND: The epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) are a first-line therapy for non-small cell lung cancer (NSCLC) with EGFR mutations. Approximately half of the patients with EGFR-mutated NSCLC are treated with EGFR-TKIs and develop disease progression within 1 year. Therefore, the early prediction of tumor progression in patients who receive EGFR-TKIs can facilitate patient management and development of treatment strategies. We proposed a deep learning approach based on both quantitative computed tomography (CT) characteristics and clinical data to predict progression-free survival (PFS) in patients with advanced NSCLC after EGFR-TKI treatment. METHODS: A total of 593 radiomic features were extracted from pretreatment chest CT images. The DeepSurv models for the progression risk stratification of EGFR-TKI treatment were proposed based on CT radiomic and clinical features from 270 stage IIIB-IV EGFR-mutant NSCLC patients. Time-dependent PFS predictions at 3, 12, 18, and 24 months and estimated personalized PFS curves were calculated using the DeepSurv models. RESULTS: The model combining clinical and radiomic features demonstrated better prediction performance than the clinical model. The model achieving areas under the curve of 0.76, 0.77, 0.76, and 0.86 can predict PFS at 3, 12, 18, and 24 months, respectively. The personalized PFS curves showed significant differences (p < 0.003) between groups with good (PFS > median) and poor (PFS < median) tumor control. CONCLUSIONS: The DeepSurv models provided reliable multi-time-point PFS predictions for EGFR-TKI treatment. The personalized PFS curves can help make accurate and individualized predictions of tumor progression. The proposed deep learning approach holds promise for improving the pre-TKI personalized management of patients with EGFR-mutated NSCLC.

Artificial Intelligence for Early Detection of Chest Nodules in X-ray Images.

Early detection increases overall survival among patients with lung cancer. This study formulated a machine learning method that processes chest X-rays (CXRs) to detect lung cancer early. After we preprocessed our dataset using monochrome and brightness correction, we used different kinds of preprocessing methods to enhance image contrast and then used U-net to perform lung segmentation. We used 559 CXRs with a single lung nodule labeled by experts to train a You Only Look Once version 4 (YOLOv4) deep-learning architecture to detect lung nodules. In a testing dataset of 100 CXRs from patients at Taipei Veterans General Hospital and 154 CXRs from the Japanese Society of Radiological Technology dataset, the sensitivity of the AI model using a combination of different preprocessing methods performed the best at 79%, with 3.04 false positives per image. We then tested the AI by using 383 sets of CXRs obtained in the past 5 years prior to lung cancer diagnoses. The median time from detection to diagnosis for radiologists assisted with AI was 46 (3-523) days, longer than that for radiologists (8 (0-263) days). The AI model can assist radiologists in the early detection of lung nodules.

Obstructive sleep apnea in young Asian adults with sleep-related complaints.

This study aimed to investigate the proportion of young OSA adults with sleep-related complaints in a sleep center, affiliated with a tertiary medical center for over a decade. This study presents a chronicle change in the numbers of young adults receiving polysomnography (PSG) and young patients with OSA from 2000 to 2017. We further analyzed 371 young patients with OSA among 2378 patients receiving PSG in our sleep center from 2016 to 2017 to capture their characteristics. Young adults constituted a substantial and relatively steady portion of examinees of PSG (25.1% ± 2.8%) and confirmed OSA cases (19.8 ± 2.4%) even though the total numbers increased with the years. Young adults with OSA tend to be sleepier, have a greater body mass index, and have a higher percentage of cigarette smoking and alcohol consumption. They also complained more about snoring and daytime sleepiness. They had a higher apnea-hypopnea index on average and experienced more hypoxemia during their sleep, both in terms of duration and the extent of desaturation. Even though the prevalence of comorbidities increased with age, hypertension in young male adults carried higher risks for OSA. Young adults with OSA have constituted a relatively constant portion of all confirmed OSA cases across time. The young adults with OSA were heavier, more symptomatic, and with more severe severity.Clinical trial: The Institutional Review Board of Taipei Veterans General Hospital approved the study (VGHIRB No. 2018-10-002CC). The study is registered with ClinicalTrials.gov, number NCT03885440.

Clinical Characteristics of Rapid Eye Movement-Related Obstructive Sleep Apnea: An Experience in a Tertiary Medical Center of Taiwan.

Purpose: Obstructive sleep apnea (OSA) is characterized by intermittent hypoxemia and sleep fragmentation. While apnea is pronounced with severe desaturation during rapid eye movement (REM) sleep, REM-related OSA is a distinct phenotype of OSA associated with respiratory disturbances predominantly during REM sleep. In this study, we investigated the clinical features of REM-related OSA in Taiwan. Patients and Methods: All patients diagnosed with OSA in the Taipei Veterans General Hospital from 2015 to 2017 were analyzed retrospectively and classified into REM-related OSA (REM-OSA) group, non-REM related OSA (NREM-OSA) group, and non-stage specific-OSA group. The clinical demographics, OSA-related symptoms, polysomnography results, and medical comorbidities of the three groups were analyzed. Results: Among 1331 patients with OSA, 414 (31.1%) were classified as REM-OSA, 808 (60.7%) as NREM-OSA, and 109 (8.2%) as non-stage specific-OSA. After being adjusted for OSA severity, the REM-OSA group was associated with less portion of males, longer desaturation duration, and lower nadir oxygen saturation (SpO2) compared with the NREM-OSA group in mild and moderate OSA. In moderate OSA, the non-stage specific-OSA group featured more OSA severity and more desaturation compared with the other groups. The Epworth Sleepiness Scale scores and the prevalence of comorbidities did not vary among the REM-OSA, NREM-OSA, and non-stage specific-OSA groups. High REM-AHI/NREM-AHI ratio was associated with young age, female gender, high BMI, and low AHI. Conclusion: OSA patients with high REM-AHI/NREM-AHI ratio are related to young age, female gender, high BMI, and low AHI. Patients with REM-related OSA presented with longer desaturation duration and lower nadir SpO2 after being adjusted for OSA severity.

Prognostic factors for advanced lung cancer patients with do-not-intubate order in intensive care unit: a retrospective study.

BACKGROUND: The survival of patients with lung cancer undergoing critical care has improved. An increasing number of patients with lung cancer have signed a predefined do-not-intubate (DNI) order before admission to the intensive care unit (ICU). These patients may still be transferred to the ICU and even receive non-invasive ventilation (NIV) support. However, there is still a lack of prognostic predictions in this cohort. Whether patients will benefit from ICU care remains unclear. METHODS: We retrospectively collected data from patients with advanced lung cancer who had signed a DNI order before ICU admission in a tertiary medical center between 2014 and 2016. The clinical characteristics and survival outcomes were discussed. RESULTS: A total of 140 patients (median age, 73 years; 62.1% were male) were included, had been diagnosed with stage III or IV non-small cell lung cancer (NSCLC) (AJCC 7th edition), and signed a DNI. Most patients received NIV during ICU stay. The median APACHE II score was 14 (standard error [SE], ± 0.66) and the mean PaO2/FiO2 ratio (P/F ratio) was 174.2 (SD, ± 104 mmHg). The APACHE II score was significantly lower in 28-day survivors (survivor: 12 (± 0.98) vs. non-survivor: 15 (± 0.83); p = 0.019). The P/F ratio of the survivors was higher than that of non-survivors (survivors: 209.6 ± 111.4 vs. non-survivors: 157.9 ± 96.7; p = 0.006). Patients with a P/F ratio ≥ 150 had better 28-day survival (p = 0.005). By combining P/F ratio ≥ 150 and APACHE II score < 16, those with high P/F ratios and low APACHE II scores during ICU admission had a notable 28-day survival compared with the rest (p < 0.001). These prognostic factors could also be applied to 90-day survival (p = 0.003). The prediction model was significant for those with driver mutations in 90-day survival (p = 0.021). CONCLUSIONS: P/F ratio ≥ 150 and APACHE II score < 16 were significant prognostic factors for critically ill patients with lung cancer and DNI. This prediction could be applied to 90-day survival in patients with driver mutations. These findings are informative for clinical practice and decision-making.

Tracheoesophageal fistula presenting with ventilator air leaks and recurrent pneumonia.

We describe a patient who received tracheostomy but complicated with tracheoesophageal fistula, where the nasogastric tube was visible from the fistula under bronchoscopy. Tracheostomy tube was then replaced with an endotracheal tube to bypass the fistula.

Application of Artificial Intelligence in Lung Cancer.

Lung cancer is the leading cause of malignancy-related mortality worldwide due to its heterogeneous features and diagnosis at a late stage. Artificial intelligence (AI) is good at handling a large volume of computational and repeated labor work and is suitable for assisting doctors in analyzing image-dominant diseases like lung cancer. Scientists have shown long-standing efforts to apply AI in lung cancer screening via CXR and chest CT since the 1960s. Several grand challenges were held to find the best AI model. Currently, the FDA have approved several AI programs in CXR and chest CT reading, which enables AI systems to take part in lung cancer detection. Following the success of AI application in the radiology field, AI was applied to digitalized whole slide imaging (WSI) annotation. Integrating with more information, like demographics and clinical data, the AI systems could play a role in decision-making by classifying EGFR mutations and PD-L1 expression. AI systems also help clinicians to estimate the patient's prognosis by predicting drug response, the tumor recurrence rate after surgery, radiotherapy response, and side effects. Though there are still some obstacles, deploying AI systems in the clinical workflow is vital for the foreseeable future.

Large-scale assessment of consistency in sleep stage scoring rules among multiple sleep centers using an interpretable machine learning algorithm.

STUDY OBJECTIVES: Polysomnography is the gold standard in identifying sleep stages; however, there are discrepancies in how technicians use the standards. Because organizing meetings to evaluate this discrepancy and/or reach a consensus among multiple sleep centers is time-consuming, we developed an artificial intelligence system to efficiently evaluate the reliability and consistency of sleep scoring and hence the sleep center quality. METHODS: An interpretable machine learning algorithm was used to evaluate the interrater reliability (IRR) of sleep stage annotation among sleep centers. The artificial intelligence system was trained to learn raters from 1 hospital and was applied to patients from the same or other hospitals. The results were compared with the experts' annotation to determine IRR. Intracenter and intercenter assessments were conducted on 679 patients without sleep apnea from 6 sleep centers in Taiwan. Centers with potential quality issues were identified by the estimated IRR. RESULTS: In the intracenter assessment, the median accuracy ranged from 80.3%-83.3%, with the exception of 1 hospital, which had an accuracy of 72.3%. In the intercenter assessment, the median accuracy ranged from 75.7%-83.3% when the 1 hospital was excluded from testing and training. The performance of the proposed method was higher for the N2, awake, and REM sleep stages than for the N1 and N3 stages. The significant IRR discrepancy of the 1 hospital suggested a quality issue. This quality issue was confirmed by the physicians in charge of the 1 hospital. CONCLUSIONS: The proposed artificial intelligence system proved effective in assessing IRR and hence the sleep center quality.

Overview of coronavirus disease 2019: Treatment updates and advances.

In late December 2019, several cases of pneumonia with unknown cause were reported in Wuhan, China, and this new type of pneumonia spread rapidly to across provinces during the subsequent weeks. The pathogen was identified quickly and was named as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The infectious disease caused by this virus is referred to as coronavirus disease 2019 (COVID-19). Within months, it has caused a global pandemic and posed a major threat to public health worldwide. As of May 23, 2020, 5 252 452 patients have been confirmed to have the disease, and 339 026 deaths have been reported. Multiple therapeutic trials are ongoing, and some promising results have been released. A vaccine would provide the most effective approach to fight the virus by preventing infection, but none are currently available. To control the COVID-19 outbreak, large-scale measures have been applied to reduce human-to-human transmission of SARS-CoV-2. Susceptible populations, including older adults, children, and healthcare providers, warrant particular attention to avoid transmission and infection. This review introduces current understanding of SARS-CoV-2 infection and treatment strategies, emphasizing the relevant challenges associated with prevention, diagnosis, and management.

Small Airway Dysfunction by Impulse Oscillometry in Symptomatic Patients with Preserved Pulmonary Function.

BACKGROUND: Asthma and chronic obstructive pulmonary disease are characterized by persistent airway inflammation and airflow limitation. Early detection of these diseases in patients with respiratory symptoms and preserved pulmonary function (PPF) defined by spirometry is difficult. Impulse oscillometry (IOS) may have better sensitivity than effort-dependent forced expiratory flow between 25% and 75% (FEF25%-75%) to detect small airway dysfunction (SAD). OBJECTIVE: To identify SAD in patients with respiratory symptoms and PPF using IOS. METHODS: Medical records of symptomatic patients without acute or known structural lung diseases were evaluated. Patients had bronchodilator testing and IOS in the outpatient clinic between March 1 and July 31, 2017. Correlations between respiratory symptoms, spirometry, and IOS parameters were determined. RESULTS: Among 349 patients enrolled to the study, 255 (73.1%) patients met the criteria of PPF. The IOS parameters-difference in resistance at 5 Hz and resistance at 20 Hz , reactance at 5 Hz, resonant frequency (Fres), and area under reactance curve between 5 Hz and resonant frequency-were significantly correlated with FEF25%-75%. The cutoffs for SAD were difference in resistance at 5 Hz and resistance at 20 Hz greater than 0.07 kPa/(L/s), reactance at 5 Hz less than -0.12 kPa/(L/s), Fres greater than 14.14 Hz, and area under reactance curve between 5 Hz and resonant frequency greater than 0.44 kPa/L. Of the IOS parameters, Fres and reactance at 5 Hz had the highest sensitivity and specificity. When compared with FEF25%-75%, Fres had greater sensitivity to detect SAD in patients with PPF. Patients with IOS-defined SAD had a significantly higher incidence of wheeze or sputum production than did those defined by FEF25%-75%. CONCLUSIONS: Patients with respiratory symptoms and PPF may have SAD, which can be identified with the aid of IOS in addition to spirometry.

Toxicities, safeties and clinical response of dacarbazine-based chemotherapy on neuroendocrine tumors in Taiwan population.

BACKGROUND: Currently, the role of dacarbazine (DTIC) based chemotherapy in neuroendocrine tumors (NETs) in Asia is unclear. Here, we report the outcomes of dacarbazine (DTIC)-based chemotherapy in Taiwan population. METHODS: DTIC alone (250 mg/m2/day), or 5-fluorouracil (5-FU, 500 mg/m2/day) and DTIC (200 mg/m2/day) with or without epirubicin (200 mg/m2/day), for 3 days, every 3-4 weeks. Subgroups were analyzed by grading, and by Ki-67 index. RESULTS: 48 patients were reviewed in this study, including 3 had grade 1 tumors, 23 had grade 2, while 22 were grade 3. In grade 3 NEC patients, the tumor Ki-67 index of 21-55% were noted in 8 patients, and >55% in 14 patients. Progression-free survival (PFS) was 5.1 months, and overall survival (OS) was 31.6 months. The PFS (in months) were 12.5 and 1.8 for patients with NETs and neuroendocrine carcinomas (NECs), respectively (p < 0.001). The OS were not reached and 5.9 months for patients with NETs and NECs, respectively (p = 0.001). Patients with NECs were divided into two groups, according to their Ki-67 index. In patients with a tumor Ki-67 index of 21-55%, PFS was 4.1 months, and OS was not reached; in those with a tumor Ki-67 index of >55%, they were 1.5 and 1.8 months, respectively (p < 0.001 and p = 0.013). CONCLUSION: NETs, and grade 3 NECs, with Ki-67 indices of 20-55% had good responses to DTIC-based chemotherapy, with acceptable side effects. Ki-67 index could predict prognosis for grade 3 NEC patients, and guide further chemotherapy choices.