Transfer Learning by Deep Tuning of Pre-trained Networks for Pulmonary Nodule Detection

Sethi, Dhaarna; Arora, Kriti; Susan, Seba

doi:10.1109/iciis51140.2020.9342686

Cited by 8 publications

(2 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During the training of both the U-Net and Attention U-Net models, we implemented both shallow tuning and deep tuning [ 41 ] of ImageNet weights. For our experiments, we proceeded with the deep tuning method because of its superior performance compared to shallow tuning of the weights, where only the final layers of the encoder model are trained as opposed to the entire encoder model.…”

Section: Discussionmentioning

confidence: 99%

Deep Learning Powered Identification of Differentiated Early Mesoderm Cells from Pluripotent Stem Cells

Mohammad,

Roy,

Karatzas

et al. 2024

Cells

View full text Add to dashboard Cite

Pluripotent stem cells can be differentiated into all three germ-layers including ecto-, endo-, and mesoderm in vitro. However, the early identification and rapid characterization of each germ-layer in response to chemical and physical induction of differentiation is limited. This is a long-standing issue for rapid and high-throughput screening to determine lineage specification efficiency. Here, we present deep learning (DL) methodologies for predicting and classifying early mesoderm cells differentiated from embryoid bodies (EBs) based on cellular and nuclear morphologies. Using a transgenic murine embryonic stem cell (mESC) line, namely OGTR1, we validated the upregulation of mesodermal genes (Brachyury (T): DsRed) in cells derived from EBs for the deep learning model training. Cells were classified into mesodermal and non-mesodermal (representing endo- and ectoderm) classes using a convolutional neural network (CNN) model called InceptionV3 which achieved a very high classification accuracy of 97% for phase images and 90% for nuclei images. In addition, we also performed image segmentation using an Attention U-Net CNN and obtained a mean intersection over union of 61% and 69% for phase-contrast and nuclear images, respectively. This work highlights the potential of integrating cell culture, imaging technologies, and deep learning methodologies in identifying lineage specification, thus contributing to the advancements in regenerative medicine. Collectively, our trained deep learning models can predict the mesoderm cells with high accuracy based on cellular and nuclear morphologies.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep Learning Powered Identification of Differentiated Early Mesoderm Cells from Pluripotent Stem Cells

Mohammad,

Roy,

Karatzas

et al. 2024

Cells

View full text Add to dashboard Cite

show abstract

“…It enables a model that has been trained for one task to be utilized for another task, usually one with less labeled data or a problem domain that is similar [19]. We used transfer learning to enhance the functionality of our CNN model for lung scan detection [20].…”

Section: Transfer Learningmentioning

confidence: 99%

Evaluation of filtering and contrast in X-ray and computerized tomography scan lung classification

Setty,

Mathad,

Shenthar

et al. 2024

IJEECS

View full text Add to dashboard Cite

Deep learning provides many convenient methods to help medical practitioners take informed decisions about diverse ailments. The goal of this project is to measure the effectiveness of filters and contrast enhancement techniques qualitatively and quantitatively in classifying lung scan images. Transfer deep learning was used to obtain the necessary results, with DenseNet 121 being the base model. Salt and pepper filter was used to introduce noise, and 3×3 mean and 5×5 mean with contrast limited adaptive histogram equalization (CLAHE) was used to minimize the effect of noise. All layers excluding the rearmost were frozen, and new dense and dropout layers were added to identify features of computerized tomography (CT) scan images of lungs. The resultant models were of comparable accuracy, where the model with no filter gave the accurate results for the given data, and the one using the 5×5 mean filter gave better adaptability in classification of unseen data. The misclassification between normal and pneumonia affected lungs is relatively higher, because of the lack of distinct features between them.

show abstract

Hybrid Deep Learning Approach with Feature Engineering for Enhanced Pulmonary Nodule Diagnosis

Bouamrane,

Derdour,

Alksas

et al. 2024

SN COMPUT. SCI.

View full text Add to dashboard Cite

Transfer Learning by Deep Tuning of Pre-trained Networks for Pulmonary Nodule Detection

Cited by 8 publications

References 26 publications

Deep Learning Powered Identification of Differentiated Early Mesoderm Cells from Pluripotent Stem Cells

Deep Learning Powered Identification of Differentiated Early Mesoderm Cells from Pluripotent Stem Cells

Evaluation of filtering and contrast in X-ray and computerized tomography scan lung classification

Hybrid Deep Learning Approach with Feature Engineering for Enhanced Pulmonary Nodule Diagnosis

Contact Info

Product

Resources

About