We present a study of subjective and objective quality assessment of compressed 4K ultra-high-definition (UHD) videos in an immersive viewing environment. First, we conduct a subjective quality ...evaluation experiment for 4K UHD videos compressed by three state-of-the-art video coding techniques, i.e., Advanced Video Coding, High Efficiency Video Coding, and VP9. In particular, we aim at investigating added values of UHD over conventional high definition (HD) in terms of perceptual quality. The results are systematically analyzed in various viewpoints, such as coding scheme, bitrate, and video content. Second, existing state-of-the-art objective quality assessment techniques are benchmarked using the subjective data in order to investigate their validity and limitation for 4K UHD videos. Finally, the video and subjective data are made publicly available for further research by the research community.
Pembrolizumab is a standard-of-care for advanced non-small-cell lung cancer (NSCLC). We assessed pembrolizumab as adjuvant therapy for completely resected stage IB–IIIA NSCLC.
In this randomised, ...triple-blind, phase 3 trial (PEARLS/KEYNOTE-091), patients were recruited from 196 medical centres in 29 countries. Eligible patients were aged 18 years or older, with completely resected, pathologically confirmed stage IB (tumours of ≥4 cm in diameter), II, or IIIA NSCLC per the American Joint Committee on Cancer staging system (7th edition) of any histology or PD-L1 expression level, and an Eastern Cooperative Oncology Group performance status of 0 or 1; adjuvant chemotherapy was to be considered for stage IB disease and was strongly recommended for stage II and IIIA disease, according to national and local guidelines. Using a central interactive voice-response system, eligible participants were randomly assigned (1:1), using a minimisation technique and stratified by disease stage, previous adjuvant chemotherapy, PD-L1 expression, and geographical region, to pembrolizumab 200 mg or placebo, both administered intravenously every 3 weeks for up to 18 cycles. Participants, investigators, and analysts were masked to treatment assignment. Dual primary endpoints were disease-free survival in the overall population and in the population with PD-L1 tumour proportion score (TPS) of 50% or greater. Efficacy was assessed in the intention-to-treat (ITT) population (ie, all participants randomly assigned to a treatment group). Safety was assessed in all participants randomly assigned to treatment who received at least one dose of study treatment. Here we report results of the second interim analysis, prespecified to occur when approximately 118 disease-free survival events had occurred in the PD-L1 TPS of 50% or greater population. This study is registered with ClinicalTrials.gov, NCT02504372, and is active but not recruiting.
Between Jan 20, 2016, and May 6, 2020, 1177 (60%) of 1955 screened participants were randomly assigned to pembrolizumab (n=590, including n=168 with PD-L1 TPS of ≥50%) or placebo (n=587; including n=165 with PD-L1 TPS of ≥50%) and included in the ITT population. Median follow-up as of data cutoff (Sept 20, 2021) for this interim analysis was 35·6 months (IQR 27·1–45·5). In the overall population, median disease-free survival was 53·6 months (95% CI 39·2 to not reached) in the pembrolizumab group versus 42·0 months (31·3 to not reached) in the placebo group (HR 0·76 95% CI 0·63–0·91, p=0·0014). In the PD-L1 TPS of 50% or greater population, median disease-free survival was not reached in either the pembrolizumab group (95% CI 44·3 to not reached) or the placebo group (95% CI 35·8 to not reached; HR 0·82 95% CI 0·57–1·18; p=0·14). Grade 3 or worse adverse events occurred in 198 (34%) of 580 participants who received pembrolizumab and 150 (26%) of 581 participants who received placebo. Grade 3 or worse events that occurred in at least ten participants in either treatment group were hypertension (35 6%) and pneumonia (12 2%) with pembrolizumab and hypertension (32 6%) with placebo. Serious adverse events occurred in 142 (24%) participants in the pembrolizumab group and 90 (15%) in the placebo group; serious adverse events that occurred in more than 1% of participants were pneumonia (13 2%), pneumonitis (12 2%), and diarrhoea (seven 1%) with pembrolizumab and pneumonia (nine 2%) with placebo. Treatment-related adverse events led to death in four (1%) participants treated with pembrolizumab (one due to both cardiogenic shock and myocarditis, one due to both septic shock and myocarditis, one due to pneumonia, and one due to sudden death) and in no participants treated with placebo.
Pembrolizumab significantly improved disease-free survival compared with placebo and was not associated with new safety signals in completely resected, PD-L1-unselected, stage IB–IIIA NSCLC. Pembrolizumab is potentially a new treatment option for stage IB–IIIA NSCLC after complete resection and, when recommended, adjuvant chemotherapy, regardless of PD-L1 expression.
Merck Sharp & Dohme, a subsidiary of Merck & Co.
Understanding music popularity is important not only for the artists who create and perform music but also for the music-related industry. It has not been studied well how music popularity can be ...defined, what its characteristics are, and whether it can be predicted, which are addressed in this paper. We first define eight popularity metrics to cover multiple aspects of popularity. Then, the analysis of each popularity metric is conducted with long-term real-world chart data to deeply understand the characteristics of music popularity in the real world. We also build classification models for predicting popularity metrics using acoustic data. In particular, we focus on evaluating features describing music complexity together with other conventional acoustic features including MPEG-7 and Mel-frequency cepstral coefficient (MFCC) features. The results show that, although room still exists for improvement, it is feasible to predict the popularity metrics of a song significantly better than random chance based on its audio signal, particularly using both the complexity and MFCC features.
Antibodies targeting the immune checkpoint molecules PD-1 or PD-L1 have demonstrated clinical efficacy in patients with metastatic non-small-cell lung cancer (NSCLC). In this trial we investigated ...the efficacy and safety of avelumab, an anti-PD-L1 antibody, in patients with NSCLC who had already received platinum-based therapy.
JAVELIN Lung 200 was a multicentre, open-label, randomised, phase 3 trial at 173 hospitals and cancer treatment centres in 31 countries. Eligible patients were aged 18 years or older and had stage IIIB or IV or recurrent NSCLC and disease progression after treatment with a platinum-containing doublet, an Eastern Cooperative Oncology Group performance status score of 0 or 1, an estimated life expectancy of more than 12 weeks, and adequate haematological, renal, and hepatic function. Participants were randomly assigned (1:1), via an interactive voice-response system with a stratified permuted block method with variable block length, to receive either avelumab 10 mg/kg every 2 weeks or docetaxel 75 mg/m2 every 3 weeks. Randomisation was stratified by PD-L1 expression (≥1% vs <1% of tumour cells), which was measured with the 73–10 assay, and histology (squamous vs non-squamous). The primary endpoint was overall survival, analysed when roughly 337 events (deaths) had occurred in the PD-L1-positive population. Efficacy was analysed in all PD-L1-positive patients (ie, PD-L1 expression in ≥1% of tumour cells) randomly assigned to study treatment (the primary analysis population) and then in all randomly assigned patients through a hierarchical testing procedure. Safety was analysed in all patients who received at least one dose of study treatment. This trial is registered with ClinicalTrials.gov, number NCT02395172. Enrolment is complete, but the trial is ongoing.
Between March 24, 2015, and Jan 23, 2017, 792 patients were enrolled and randomly assigned to receive avelumab (n=396) or docetaxel (n=396). 264 participants in the avelumab group and 265 in the docetaxel group had PD-L1-positive tumours. In patients with PD-L1-positive tumours, median overall survival did not differ significantly between the avelumab and docetaxel groups (11·4 months 95% CI 9·4–13·9 vs 10·3 months 8·5–13·0; hazard ratio 0·90 96% CI 0·72–1·12; one-sided p=0·16). Treatment-related adverse events occurred in 251 (64%) of 393 avelumab-treated patients and 313 (86%) of 365 docetaxel-treated patients, including grade 3–5 events in 39 (10%) and 180 (49%) patients, respectively. The most common grade 3–5 treatment-related adverse events were infusion-related reaction (six patients 2%) and increased lipase (four 1%) in the avelumab group and neutropenia (51 14%), febrile neutropenia (37 10%), and decreased neutrophil counts (36 10%) in the docetaxel group. Serious treatment-related adverse events occurred in 34 (9%) patients in the avelumab group and 75 (21%) in the docetaxel group. Treatment-related deaths occurred in four (1%) participants in the avelumab group, two due to interstitial lung disease, one due to acute kidney injury, and one due to a combination of autoimmune myocarditis, acute cardiac failure, and respiratory failure. Treatment-related deaths occurred in 14 (4%) patients in the docetaxel group, three due to pneumonia, and one each due to febrile neutropenia, septic shock, febrile neutropenia with septic shock, acute respiratory failure, cardiovascular insufficiency, renal impairment, leucopenia with mucosal inflammation and pyrexia, infection, neutropenic infection, dehydration, and unknown causes.
Compared with docetaxel, avelumab did not improve overall survival in patients with platinum-treated PD-L1-positive NSCLC, but had a favourable safety profile.
Merck and Pfizer.
Imbalanced data classification is a challenging problem frequently encountered in many real-world applications. Traditional classification algorithms are generally designed to maximize overall ...accuracy; therefore, their effectiveness tends to be impeded by imbalanced data. Similar to other traditional classifiers, naive Bayes (NB) sometimes fails at predicting minority instances owing to its sensitivity to class distribution. To cope with this challenge, we proposed RankOptAUC NB (RNB), a novel attribute weighting method for the NB. In the proposed method, learning a weighted NB classifier was formulated as a nonlinear optimization problem with the objective of maximizing the area under the ROC (AUC). The optimization formulation enabled the RNB method to select important variables by simply adding a regularization term to the objective function. We also provided theoretical evidence that, based on the AUC metric, the proposed method improved the performance of a weighted NB classifier. The results of numerical experiments conducted using 30 real-world datasets proved that the proposed scheme successfully determined the optimal attribute weights for imbalanced data classification.
•A novel weighted naive Bayes (NB) for imbalanced data classification was proposed.•Learning a weighted NB classifier was formulated as a nonlinear optimization problem.•Area under ROC curve (AUC) was incorporated into the objective function.•The proposed method can select important attributes.
Tumor spread through air spaces (STAS) is an invasive pattern of lung cancer that was recently described. In this study, we investigated the association between the extent of STAS and ...clinicopathological characteristics and patient outcomes in resected non-small cell lung cancers (NSCLCs). STAS has been prospectively described from 2008 and graded its extent with a two-tiered system (STAS I: <2500 μm one field of ×10 objective lens from the edge of tumor and STAS II: ≥2500 μm from the edge of tumor) from 2011 in Seoul National University Bundang Hospital. We retrospectively analyzed the correlations between the extent of STAS and clinicopathologic characteristics and prognostic significance in 1869 resected NSCLCs. STAS was observed in 765 cases (40.9%) with 456 STAS I (24.4%) and 309 STAS II (16.5%). STAS was more frequently found in patients with adenocarcinoma (ADC) (than squamous cell carcinoma), pleural invasion, lymphovascular invasion, and/or higher pathologic stage. In ADC, there were significant differences in recurrence free survival (RFS), overall survival (OS), and lung cancer specific survival (LCSS) according to the extent of STAS. In stage IA non-mucinous ADC, multivariate analysis revealed that STAS II was significantly associated with shorter RFS and LCSS (p < 0.001 and p = 0.006, respectively). In addition, STAS II was an independent poor prognostic factor for recurrence in both limited and radical resection groups (p = 0.001 and p = 0.023, respectively). In conclusion, presence of STAS II was an independent poor prognostic factor in stage IA non-mucinous ADC regardless of the extent of resection.
The purpose of this study is to develop a prediction model that identifies the potential risk of fatality accidents at construction sites using machine learning based on industrial accident data ...collected by the Ministry of Employment and Labor (MOEL) of the Republic of Korea from 2011 to 2016. The data details 137,323 injuries and 2846 deaths, and includes age, sex, and length of service of each accident victim, as well as the type of construction, employer scale, and date of the accident. Upon describing the distribution of the dataset, machine learning methods, such as logistic regression, decision tree, random forest, and AdaBoost analyses were applied with the derivation of major variables influencing classification in each algorithm. A comparison of the performance of each model showed the area under the receiver operating characteristic (AUROC) curve to be highest for the random forest method, at 0.9198, which translates to a 91.98% successful predictive rate in terms of classifying workers who could face a high fatality risk. The random forest analysis of this study indicates that the month (season) and employment size are the most influential factors, followed by age, weekday, and service length based on mean decrease Gini values to predict the likelihood of a fatality accident. Moreover, this analysis generated ensemble predictions based on all the factors contained in the dataset. Hence, this study demonstrates the feasibility of machine learning in the construction safety management area. The results obtained can contribute to the prevention of accidents by raising awareness of potential safety risks, by quantitatively predicting fatal accidents and incorporating the findings with a manpower control system at a construction site.
•A predictive model for the likelihood of a fatality accident was compared and developed using machine learning methods.•The random forest method showed a predictive rate of 91.98% success at classifying workers who might face a fatality risk.•The season and employment size are the most influential factors in the model, followed by age, weekday, and service length.•The prediction model is expected to prevent accidents by raising awareness of potential safety risks.
Patients with EGFR-mutated non-small-cell lung cancer (NSCLC) given EGFR tyrosine kinase inhibitors (TKIs) inevitably become resistant to first-generation or second-generation drugs. We assessed the ...safety, tolerability, pharmacokinetics, and activity of lazertinib—an irreversible, third-generation, mutant-selective, EGFR TKI—in patients with advanced NSCLC progressing after EGFR TKI therapy.
This first-in-human, open-label, multicentre, phase 1–2 study had three parts: dose escalation, dose expansion, and dose extension; here, we report results on dose escalation and dose expansion. The study was done in 14 hospitals in Korea. Eligible patients were aged 20 years or older and had advanced NSCLC harbouring an activating EGFR mutation and progressing after first-generation or second-generation EGFR TKI treatment, a defined tumour T790M mutation status, an Eastern Cooperative Oncology Group performance status of 0–1, at least one measurable extracranial lesion, defined according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, and adequate organ function. Patients were enrolled to seven dose-escalation cohorts according to a rolling six design; five cohorts were expanded. Patients were given oral lazertinib 20 mg, 40 mg, 80 mg, 120 mg, 160 mg, 240 mg, or 320 mg once daily continuously in 21-day cycles. Primary endpoints were safety and tolerability and secondary endpoints included objective response in evaluable patients. This study is registered with ClinicalTrials.gov, NCT03046992, and the phase 2 extension study is ongoing.
Between Feb 15, 2017, and May 28, 2018, 127 patients were enrolled into the dose escalation group (n=38) and dose expansion group (n=89). No dose-limiting toxicities occurred. There was no dose-dependent increase in adverse events. The most commonly reported adverse events were grade 1–2 rash or acne (in 38 30% of 127 patients) and pruritus (in 34 27%). Grade 3 or grade 4 adverse events occurred in 20 (16%) patients, with the most common being grade 3 pneumonia (four 3%). Treatment-related grade 3 or 4 adverse events occurred in four (3%) patients; treatment-related serious adverse events were reported in six patients (5%). There were no adverse events with an outcome of death and no treatment-related deaths. The proportion of patients achieving an objective response by independent central review assessment was 69 (54%; 95% CI 46–63) of 127.
Lazertinib had a tolerable safety profile and showed promosing clinical activity in patients with NSCLC progressing on or after EGFR TKI therapy. Our findings provide a rationale for further clinical investigations.
Yuhan Corporation.
Non-small-cell lung cancer (NSCLC) with epidermal growth factor receptor (
) exon 20 insertion (Exon20ins) mutations exhibits inherent resistance to approved tyrosine kinase inhibitors. Amivantamab, ...an EGFR-MET bispecific antibody with immune cell-directing activity, binds to each receptor's extracellular domain, bypassing resistance at the tyrosine kinase inhibitor binding site.
CHRYSALIS is a phase I, open-label, dose-escalation, and dose-expansion study, which included a population with
Exon20ins NSCLC. The primary end points were dose-limiting toxicity and overall response rate. We report findings from the postplatinum
Exon20ins NSCLC population treated at the recommended phase II dose of 1,050 mg amivantamab (1,400 mg, ≥ 80 kg) given once weekly for the first 4 weeks and then once every 2 weeks starting at week 5.
In the efficacy population (n = 81), the median age was 62 years (range, 42-84 years); 40 patients (49%) were Asian, and the median number of previous lines of therapy was two (range, 1-7). The overall response rate was 40% (95% CI, 29 to 51), including three complete responses, with a median duration of response of 11.1 months (95% CI, 6.9 to not reached). The median progression-free survival was 8.3 months (95% CI, 6.5 to 10.9). In the safety population (n = 114), the most common adverse events were rash in 98 patients (86%), infusion-related reactions in 75 (66%), and paronychia in 51 (45%). The most common grade 3-4 adverse events were hypokalemia in six patients (5%) and rash, pulmonary embolism, diarrhea, and neutropenia in four (4%) each. Treatment-related dose reductions and discontinuations were reported in 13% and 4% of patients, respectively.
Amivantamab, via its novel mechanism of action, yielded robust and durable responses with tolerable safety in patients with
Exon20ins mutations after progression on platinum-based chemotherapy.
•A novel deep learning multimodal classification architecture is proposed.•It supports high compatibility with existing deep learning architectures.•It thoroughly considers correlated information ...between different modalities.•It ensures robustness against limited availability of data.
Classification using multimodal data arises in many machine learning applications. It is crucial not only to model cross-modal relationship effectively but also to ensure robustness against loss of part of data or modalities. In this paper, we propose a novel deep learning-based multimodal fusion architecture for classification tasks, which guarantees compatibility with any kind of learning models, deals with cross-modal information carefully, and prevents performance degradation due to partial absence of data. We employ two datasets for multimodal classification tasks, build models based on our architecture and other state-of-the-art models, and analyze their performance on various situations. The results show that our architecture outperforms the other multimodal fusion architectures when some parts of data are not available.