Deep Learning for the Classification of Small-Cell and Non-Small-Cell Lung Cancer

被引：69

作者：

Kriegsmann, Mark ^{[1
,2
]}

Haag, Christian ^{[1
,3
]}

Weis, Cleo-Aron ^{[4
]}

Steinbuss, Georg ^{[1
,3
]}

Warth, Arne ^{[5
]}

Zgorzelski, Christiane ^{[1
]}

Muley, Thomas ^{[2
,6
]}

Winter, Hauke ^{[2
,6
]}

Eichhorn, Martin E. ^{[2
,6
]}

Eichhorn, Florian ^{[2
,6
]}

Kriegsmann, Joerg ^{[7
,8
]}

Christopolous, Petros ^{[2
,9
]}

Thomas, Michael ^{[2
,9
]}

Witzens-Harig, Mathias ^{[10
]}

Sinn, Peter ^{[1
]}

von Winterfeld, Moritz ^{[1
]}

Heussel, Claus Peter ^{[2
,11
,12
]}

Herth, Felix J. F. ^{[2
,13
]}

Klauschen, Frederick ^{[14
]}

Stenzinger, Albrecht ^{[1
,2
]}

Kriegsmann, Katharina ^{[3
]}

机构：

[1] Heidelberg Univ, Inst Pathol, D-69120 Heidelberg, Germany

[2] German Ctr Lung Res DZL, Translat Lung Res Ctr Heidelberg, D-69120 Heidelberg, Germany

[3] Heidelberg Univ, Dept Hematol Oncol & Rheumatol, D-69120 Heidelberg, Germany

[4] Heidelberg Univ, Univ Med Ctr Mannheim, Inst Pathol, D-68782 Mannheim, Germany

[5] UEGP MVZ Giessen Wetzlar Limburg, Inst Pathol Cytopathol & Mol Pathol, D-65549 Limburg, Germany

[6] Heidelberg Univ, Dept Thorac Surg, Thoraxklin, D-69126 Heidelberg, Germany

[7] Mol Pathol Trier, D-54296 Trier, Germany

[8] Danube Private Univ Krems, A-3500 Krems, Austria

[9] Heidelberg Univ, Dept Thorac Oncol, Thoraxklin, D-69126 Heidelberg, Germany

[10] Heidelberg Univ, Med Fac, D-69120 Heidelberg, Germany

[11] Heidelberg Univ, Dept Diagnost & Intervent Radiol Nucl Med, Thoraxklin, D-69120 Heidelberg, Germany

[12] Heidelberg Univ, Dept Diagnost & Intervent Radiol, Thoraxklin, D-69120 Heidelberg, Germany

[13] Heidelberg Univ, Dept Pneumol & Crit Care Med, Thoraxklin, D-69126 Heidelberg, Germany

[14] Univ Hosp Charite, Inst Pathol, D-10117 Berlin, Germany

来源：

CANCERS | 2020年 / 12卷 / 06期

关键词：

artificial intelligence; deep learning; lung cancer; histology; non-small cell lung cancer; small cell lung cancer; DIGITAL PATHOLOGY; PULMONARY; IMMUNOMARKERS; DIAGNOSIS;

D O I：

10.3390/cancers12061604

中图分类号：

R73 [肿瘤学];

学科分类号：

100214 ;

摘要：

Reliable entity subtyping is paramount for therapy stratification in lung cancer. Morphological evaluation remains the basis for entity subtyping and directs the application of additional methods such as immunohistochemistry (IHC). The decision of whether to perform IHC for subtyping is subjective, and access to IHC is not available worldwide. Thus, the application of additional methods to support morphological entity subtyping is desirable. Therefore, the ability of convolutional neuronal networks (CNNs) to classify the most common lung cancer subtypes, pulmonary adenocarcinoma (ADC), pulmonary squamous cell carcinoma (SqCC), and small-cell lung cancer (SCLC), was evaluated. A cohort of 80 ADC, 80 SqCC, 80 SCLC, and 30 skeletal muscle specimens was assembled; slides were scanned; tumor areas were annotated; image patches were extracted; and cases were randomly assigned to a training, validation or test set. Multiple CNN architectures (VGG16, InceptionV3, and InceptionResNetV2) were trained and optimized to classify the four entities. A quality control (QC) metric was established. An optimized InceptionV3 CNN architecture yielded the highest classification accuracy and was used for the classification of the test set. Image patch and patient-based CNN classification results were 95% and 100% in the test set after the application of strict QC. Misclassified cases mainly included ADC and SqCC. The QC metric identified cases that needed further IHC for definite entity subtyping. The study highlights the potential and limitations of CNN image classification models for tumor differentiation.

引用

页码：1 / 15

页数：16

共 59 条

[21]

Guo ZC, 2019, SCI REP-UK, V9, DOI [10.1038/s41598-018-37492-9, 10.1038/s41598-019-38952-6]

[22] Applications and challenges of digital pathology and whole slide imaging [J].