scholarly journals Classification and Stage Prediction of Lung Cancer using Convolutional Neural Networks

2019 ◽  
Vol 8 (10) ◽  
pp. 993-998
Keyword(s):  
Lung Cancer ◽  
State Of The Art ◽  
Early Stage ◽  
Signs And Symptoms ◽  
Large Cell ◽  
Training Data ◽  
Large Set ◽  
Fine Grained ◽  
Training Samples ◽  
Large Cell Lung Cancer

In recent years, digital image processing is widely used for the medical treatment classification and diagnosis. Lung cancer is the most leading cause of death in all over the world nowadays. Based on the signs and symptoms it can’t be diagnosis and treatment classified at the early stage. However it can be identified through the symptoms like coughing up blood and chest pain, the stages and risk factors of the cancer cannot be identified through the symptoms. The CT scanned lung images should be involved in image classification processing for earlier prediction of stages and treatment diagnosis. In existing, machine learning treatment classification can be done through the SVM classification. In case of large set of training samples, this will not be in accurate manner and it has less accuracy because of improper feature extraction techniques. Thus the performance of the classification based on the segmented features obtained in preceding sections. The extracted fine-grained training data through deep learning are utilized for the classification using Convolution Neural Network (CNN). In this paper, we propose a novel framework to classify both small cell and large cell lung cancer and predict its type and treatment using CNN. It is also concentrates on the preprocessing and segmentation processes to accomplish the accuracy in prediction. The experiment results in Python - TensorFlow with Kaggle image dataset show that compared to state of the art of classification and prediction methods, the proposed scheme can obtain much higher accuracy in type prediction and treatment diagnosis.

scholarly journals New polyp image classification technique using transfer learning of network-in-network structure in endoscopic images

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Young Jae Kim ◽  
Jang Pyo Bae ◽  
Jun-Won Chung ◽  
Dong Kyun Park ◽  
Kwang Gi Kim ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Transfer Learning ◽  
Test Data ◽  
State Of The Art ◽  
Early Stage ◽  
Statistical Significance ◽  
Recall Rate ◽  
Training Data ◽  
Fine Tuning ◽  
Accuracy Evaluation

AbstractWhile colorectal cancer is known to occur in the gastrointestinal tract. It is the third most common form of cancer of 27 major types of cancer in South Korea and worldwide. Colorectal polyps are known to increase the potential of developing colorectal cancer. Detected polyps need to be resected to reduce the risk of developing cancer. This research improved the performance of polyp classification through the fine-tuning of Network-in-Network (NIN) after applying a pre-trained model of the ImageNet database. Random shuffling is performed 20 times on 1000 colonoscopy images. Each set of data are divided into 800 images of training data and 200 images of test data. An accuracy evaluation is performed on 200 images of test data in 20 experiments. Three compared methods were constructed from AlexNet by transferring the weights trained by three different state-of-the-art databases. A normal AlexNet based method without transfer learning was also compared. The accuracy of the proposed method was higher in statistical significance than the accuracy of four other state-of-the-art methods, and showed an 18.9% improvement over the normal AlexNet based method. The area under the curve was approximately 0.930 ± 0.020, and the recall rate was 0.929 ± 0.029. An automatic algorithm can assist endoscopists in identifying polyps that are adenomatous by considering a high recall rate and accuracy. This system can enable the timely resection of polyps at an early stage.

scholarly journals BeautyNet: Joint Multiscale CNN and Transfer Learning Method for Unconstrained Facial Beauty Prediction

2019 ◽  
Vol 2019 ◽  
pp. 1-14 ◽  
Author(s):  
Yikui Zhai ◽  
He Cao ◽  
Wenbo Deng ◽  
Junying Gan ◽  
Vincenzo Piuri ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Classification Accuracy ◽  
Learning Strategy ◽  
State Of The Art ◽  
Activation Function ◽  
Training Data ◽  
Fine Grained ◽  
Pattern Recognition Problem ◽  
Face Features ◽  
Facial Beauty

Because of the lack of discriminative face representations and scarcity of labeled training data, facial beauty prediction (FBP), which aims at assessing facial attractiveness automatically, has become a challenging pattern recognition problem. Inspired by recent promising work on fine-grained image classification using the multiscale architecture to extend the diversity of deep features, BeautyNet for unconstrained facial beauty prediction is proposed in this paper. Firstly, a multiscale network is adopted to improve the discriminative of face features. Secondly, to alleviate the computational burden of the multiscale architecture, MFM (max-feature-map) is utilized as an activation function which can not only lighten the network and speed network convergence but also benefit the performance. Finally, transfer learning strategy is introduced here to mitigate the overfitting phenomenon which is caused by the scarcity of labeled facial beauty samples and improves the proposed BeautyNet’s performance. Extensive experiments performed on LSFBD demonstrate that the proposed scheme outperforms the state-of-the-art methods, which can achieve 67.48% classification accuracy.

scholarly journals MA23.07 Defining the Role of Adjuvant Therapy for Early Stage Large Cell Neuroendocrine Lung Cancer

2018 ◽  
Vol 13 (10) ◽  
pp. S440
Author(s):  
E. Wakeam ◽  
S. Stokes ◽  
A. Adibfar ◽  
N. Leighl ◽  
M. Giuliani ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Adjuvant Therapy ◽  
Early Stage ◽  
Large Cell

scholarly journals Unsupervised Deep Learning via Affinity Diffusion

2020 ◽  
Vol 34 (07) ◽  
pp. 11029-11036
Author(s):  
Jiabo Huang ◽  
Qi Dong ◽  
Shaogang Gong ◽  
Xiatian Zhu
Keyword(s):  
Deep Learning ◽  
State Of The Art ◽  
General Purpose ◽  
Training Data ◽  
Learning Approach ◽  
Model Learning ◽  
Feature Representations ◽  
Discriminative Feature ◽  
Training Samples ◽  
Unsupervised Deep Learning

Convolutional neural networks (CNNs) have achieved unprecedented success in a variety of computer vision tasks. However, they usually rely on supervised model learning with the need for massive labelled training data, limiting dramatically their usability and deployability in real-world scenarios without any labelling budget. In this work, we introduce a general-purpose unsupervised deep learning approach to deriving discriminative feature representations. It is based on self-discovering semantically consistent groups of unlabelled training samples with the same class concepts through a progressive affinity diffusion process. Extensive experiments on object image classification and clustering show the performance superiority of the proposed method over the state-of-the-art unsupervised learning models using six common image recognition benchmarks including MNIST, SVHN, STL10, CIFAR10, CIFAR100 and ImageNet.

scholarly journals A CASE OF RECURRENT LARGE CELL LUNG CANCER PRESENTING WITH ILEUS

2008 ◽  
Vol 69 (6) ◽  
pp. 1341-1344 ◽  
Author(s):  
Motohiro NISHIMURA ◽  
Masao KOBAYASHI ◽  
Kanji KAWAI ◽  
Yasuhiko TANIOKA ◽  
Kenichiro HAMAGASHIRA ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Cell Lung Cancer ◽  
Large Cell ◽  
Large Cell Lung Cancer

scholarly journals Relevance-guided Supervision for OpenQA with ColBERT

2021 ◽  
Vol 9 ◽  
pp. 929-944
Author(s):  
Omar Khattab ◽  
Christopher Potts ◽  
Matei Zaharia
Keyword(s):  
Question Answering ◽  
State Of The Art ◽  
Training Data ◽  
Coarse Grained ◽  
Retrieval Model ◽  
Open Domain ◽  
Weak Supervision ◽  
Fine Grained ◽  
Vector Representations ◽  
Large Corpus

Abstract Systems for Open-Domain Question Answering (OpenQA) generally depend on a retriever for finding candidate passages in a large corpus and a reader for extracting answers from those passages. In much recent work, the retriever is a learned component that uses coarse-grained vector representations of questions and passages. We argue that this modeling choice is insufficiently expressive for dealing with the complexity of natural language questions. To address this, we define ColBERT-QA, which adapts the scalable neural retrieval model ColBERT to OpenQA. ColBERT creates fine-grained interactions between questions and passages. We propose an efficient weak supervision strategy that iteratively uses ColBERT to create its own training data. This greatly improves OpenQA retrieval on Natural Questions, SQuAD, and TriviaQA, and the resulting system attains state-of-the-art extractive OpenQA performance on all three datasets.

scholarly journals Case Report: Long-Term Survival With Anlotinib in a Patient With Advanced Undifferentiated Large-Cell Lung Cancer and Rare Tonsillar Metastasis

2021 ◽  
Vol 11 ◽  
Author(s):  
Tianwei Xu ◽  
Chenchen Wei ◽  
Xiaoteng Zou ◽  
Binbin Lu ◽  
Zhaoxia Wang
Keyword(s):  
Lung Cancer ◽  
Cell Lung Cancer ◽  
Case Reports ◽  
Antiangiogenic Therapy ◽  
Local Treatment ◽  
Chemotherapy Resistance ◽  
Large Cell ◽  
Term Survival ◽  
Good Efficacy ◽  
Large Cell Lung Cancer

Undifferentiated large-cell lung cancer is a rare type of non-small cell lung cancer (NSCLC) with a poor prognosis. It is insensitive to chemotherapy and easily develops drug resistance. Analysis of the Surveillance, Epidemiology, and End Results (SEER) database showed that patients with stage IV undifferentiated large-cell lung cancer had a median overall survival (OS) of only 4 months and that those who received chemotherapy had a median OS of only 5 months longer than those who did not. For the first time, we report a case of advanced large-cell undifferentiated lung cancer with rare tonsil metastasis. The patient developed resistance after 3 months of platinum-based systemic chemotherapy and local treatment. Antiangiogenic therapy has been continuously progressing and has shown certain efficacy in treating many malignant tumors, such as lung cancer. However, there are no relevant studies or case reports on antiangiogenic therapy in the treatment of undifferentiated large-cell lung cancer. Anlotinib, an orally delivered small-molecule antiangiogenic tyrosine kinase inhibitor (TKI), was administered to this patient after chemotherapy resistance occurred, and the outcome was assessed as continued stable disease (SD). As of the last follow-up evaluation, the progression-free survival (PFS) of the patient was 21.5 months, and the OS was 27.5 months. Retrospective immunohistochemical analysis showed that the patient was positive for one of the targets of anlotinib (PDGFR). In general, the findings in this case suggest that anlotinib may be an option with good efficacy for patients with large-cell undifferentiated lung cancer after chemotherapy resistance that may have good efficacy and also suggest that PDGFR may be the target underlying this effect.

Patterns of lung cancer in people living with human immunodeficiency virus (HIV): A retrospective, single-center study in Washington, D.C.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. e21154-e21154
Author(s):  
Margaret Pruitt ◽  
Rajesh Naidu Janapala ◽  
Faysal Haroun
Keyword(s):  
Lung Cancer ◽  
Late Stage ◽  
Early Stage ◽  
Primary Lung Cancer ◽  
Large Cell ◽  
Molecular Data ◽  
Stage I ◽  
Stage Iii ◽  
Small Cell Lung ◽  
Neuroendocrine Cancer

e21154 Background: Lung cancer is the leading cause of cancer death and the most common non-acquired immune deficiency syndrome defining malignancy in people living with HIV (PLWH). Disparities in outcomes have been observed despite lung cancer mortality reportedly decreasing in the general population over the last decade due to lower rates of smoking and the advent of novel therapies. To better understand the current trend in lung cancer in PLWH, we explored demographic characteristics, comorbidities, and lung cancer pathology and molecular data in this population. Methods: A retrospective search of patient charts was conducted from 2004 to January 2021 using billing codes for HIV and primary lung cancer. Patients who had incorrect HIV or primary lung cancer diagnoses were excluded. Results: The search yielded 45 patients, of which 11 were excluded as described above: 66% were males, 82% African American, and 18% Caucasian. About two-thirds of patients were living in zip codes with predominantly low to medium household incomes. The median pack years of patients diagnosed with Stage I or II non-small cell lung cancer (NSCLC) was 40, Stage III or IV NSCLC was 20, early stage small cell lung cancer (SCLC) was 30, and late stage SCLC was 60. The median time between HIV and lung cancer diagnoses was 21.7 years for Stage I or II NSCLC, 17.1 years for Stage III or IV NSCLC, 15.2 for early stage SCLC, and 13.3 for late stage SCLC. Of 26 patients with viral load (VL) data, 21 (80.7%) had VL less than 500 when lung cancer was diagnosed. Of the 33 charts with available pathology data, there were 16 adenocarcinomas, 6 squamous carcinomas, 3 adenosquamous carcinomas, 1 large cell neuroendocrine cancer, 4 SCLCs, 1 mesothelioma, and 2 unspecified NSCLCs. Of 19 patients with a histologic grade, 11 had a high-grade tumor (57.9%). For the NSCLCs, 8 were Stage I (28.5%), 2 Stage II (7.1%), 8 Stage III (28.5%), 9 Stage IV (32.1%), and 1 with an unspecified stage. One SCLC was early stage and the remaining 3 were late stage. Five patients had brain metastasis. Molecular data or PDL-1 expression was available for 10 adenocarcinomas (62.5%), 1 adenosquamous (33%), 3 squamous carcinomas (50%), and the large cell neuroendocrine cancer. An EGFR mutation was detected in 2 cancers. ALK rearrangement was found in 1. Other mutations were detected. Two cancers were in each PDL1 expression category: < 1%, 1-50%, and > 50%. Conclusions: Our study suggests that PLWH with lung cancer continue to have high rates of smoking. Viral load was well controlled. A range in stages of lung cancer was observed including earlier stages. Although molecular data was limited, available EGFR and ALK gene alterations, and PD-L1 expression prevalence were on par with that of the general population. With advancements in lung cancer treatment, additional research is needed in the PLWH population to better understand and mitigate disparities.

Sign in / Sign up

Export Citation Format

Share Document