PET/CT-Based Image Analysis and Machine Learning of Hypermetabolic Pulmonary Lesions

PET/CT Imaging-Based Distinction of Pulmonary Lymphoma and Other Hypermetabolic Lesions Via Imaging Manifestations and Machine Learning Techniques: a Multicenter Retrospective Study

First, we analyse the types, imaging findings and relevant treatment responses based on PET/CT to complete a more comprehensive view of pulmonary lymphomas.

Then, some models based on radiomics features will be developed to verify the possibility of differentiating pulmonary lymphomas via machine learning and develop a multi-class classification model.

The final objective of this study is to develop a set of deep learning models for preliminary lung lesion segmentation and multi-class classification. The models will classify FDG-avid lung lesions into four groups, each defined by their pathological origin, primary therapy and relevant clinical department.

Study Overview

• Status: Completed
• Conditions: Pulmonary Metastases; Lung Cancers; Pulmonary Lymphomas; Benign Pulmonary Diseases
• Intervention / Treatment: Other: Observe the medical images; Other: Feature extraction

Detailed Description

1. The local image feature extraction software (LIFEx, v 7.4.0, France) was employed for the image review and measurement of relevant data. Three observers independently interpreted the images. In cases of disagreement, the opinion of a senior doctor with over a decade of experience was given precedence. The imaging findings were recorded based on the baseline examinations. Lesion counts, locations, and descriptive labels were systematically logged in accordance with the norms set out in imaging report. The statistical software SPSS (v26.0) was used in data sorting and calculation. Chi-square test was employed to compare SPL and PPL based on categorical variables like CT findings, while T-test was used to assess continuous variables like glycemia and SUV. Given the predominance of categorical variables, chi-square, or Fisher's exact test (for samples <40 or >20% cells with <5 expected counts) was utilised to assess treatment response and imaging performance. Spearman's correlation coefficient was employed to analyse the relationship between categorical and SUV-based continuous variables.
2. In this study, the metabolic tumor volume at a relative threshold of 40% (MTV40%) was selected as the volume of interest (VOI) for image analysis. For feature extraction, we employed the Python (v3.11.7)-based radiomics feature extraction toolkit PyRadiomics (v3.1.0), along with the medical image processing library SimpleITK (v2.3.1), the numerical computation and data manipulation library Numpy (v1.26.2), and the wavelet transform library PyWavelet (v1.5.0). Feature selection was conducted using RStudio (v.2023.12.0+369) based on the R programming language (v4.2.0). To ensure computational efficiency and avoid overfitting, the number of features retained was limited to 10% or less of the number of lesions in the training set. Model analysis and validation were primarily performed using RStudio as well.
3. The deep learning study divides the task of identifying and classifying hypermetabolic lung lesions into two stages: segmentation and classification. In the segmentation stage, we first utilized the open-source 2D model Lungmask to automatically crop the lung region from whole-body PET/CT images, ensuring that subsequent processing is focused on the lung area. Next, we developed a 3D UNet model with residual modules specifically designed for segmenting hypermetabolic lung lesions. This model takes the cropped PET/CT images as input, efficiently extracting lesion information from the three-dimensional images and accurately segmenting the hypermetabolic lung lesion areas.The model was then applied to both internal test sets and external validation sets for inference, resulting in the extraction of lesion-containing ROIs.

• Study Type: Observational
• Enrollment (Actual): 647

Contacts and Locations

Study Locations

• China
  • Shanghai
    • Shanghai, Shanghai, China, 200025
      • Ruijin Hospital affiliated to Shanghai Jiao Tong University of Medicine

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

This retrospective study enrolled patients who underwent PET/CT exams at the Nuclear Medicine Department from January 2015 to Feburary 2024 in 5 institutions: Ruijin Hospital, Proton Center of Ruijin North Hospital, Shanghai Pulmonary Hospital, Lu'an People's Hospital of Anhui Province and the Affiliated Hospital of Nanjing University of Traditional Chinese Medicine.

Description

Inclusion criteria:

1. Adult patients (≥18 years);
2. Primary or recurrent lymphoma, ≥6 months from last treatment; primary lung cancer patients without prior malignancy;
3. Benign solid lung lesions, without prior malignancy;
4. Pulmonary metastasis, untreated with lung radiotherapy or particle implantation;
5. Baseline assessment revealing PET-positive pulmonary lesions.
6. Pathological results within 3 months of exam date, confirmed lung lesion types via tracheoscopy, lung puncture, or surgery.
7. Baseline pulmonary lesions remaining considered to be pulmonary lymphoma (or metastases) based on follow-up clinical and imaging evaluation.

Exclusion criteria:

1. Poor image quality;
2. Inability to delineate the boundaries of lung lesions on CT images;
3. Artifacts caused by nearby devices such as stents or drainage tubes.

Study Plan

How is the study designed?

Number of groups / cohorts

4

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Pulmonary lymphoma; (1) Adult patients (≥18 years). (2) Patients with primary or recurrent lymphoma, ≥6 months from last treatment. (3) Baseline assessment at hospital revealed PET-positive pulmonary lesions, CT-measured maximum diameter ≥3mm, visible across ≥2 image layers. (4) Pathological results within 3 months of exam date, confirmed lung lesion types via tracheoscopy, lung puncture, or surgery. Or baseline pulmonary lesions of lymphoma diagnosed by lymph node and external lung puncture, remains considered to be pulmonary lymphoma based on follow-up clinical and imaging evaluation.	Other: Observe the medical images; Observe the medical images via work station or local image analysing software; Other: Feature extraction; Extracting image feature via radiomics or deep learning methods
Lung cancer; (1) Adult patients (≥18 years). (2) Patients with primary lung cancer patients without prior malignancy (3) Baseline assessment at hospital revealed PET-positive pulmonary lesions, CT-measured maximum diameter ≥3mm, visible across ≥2 image layers. (4) Pathological results within 3 months of exam date, confirmed lung lesion types via tracheoscopy, lung puncture, or surgery.	Other: Observe the medical images; Observe the medical images via work station or local image analysing software; Other: Feature extraction; Extracting image feature via radiomics or deep learning methods
Benign; (1) Adult patients (≥18 years). (2) Patients with benign solid lung lesions, without prior malignancy. (3) Baseline assessment at hospital revealed PET-positive pulmonary lesions, CT-measured maximum diameter ≥3mm, visible across ≥2 image layers. (4) Pathological results within 3 months of exam date, confirmed lung lesion types via tracheoscopy, lung puncture, or surgery.	Other: Observe the medical images; Observe the medical images via work station or local image analysing software; Other: Feature extraction; Extracting image feature via radiomics or deep learning methods
Metastasis; (1) Adult patients (≥18 years). (2) Pulmonary metastatic patients, untreated with lung radiotherapy or particle implantation. (3) Baseline assessment at hospital revealed PET-positive pulmonary lesions, CT-measured maximum diameter ≥3mm, visible across ≥2 image layers. (4) Pathological results within 3 months of exam date, confirmed lung lesion types via tracheoscopy, lung puncture, or surgery. Or baseline pulmonary lesions of metastases diagnosed by lymph node and external lung puncture, remains considered to be pulmonary metastases based on follow-up clinical and imaging evaluation.	Other: Observe the medical images; Observe the medical images via work station or local image analysing software; Other: Feature extraction; Extracting image feature via radiomics or deep learning methods

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Time Frame
Imaging/radiomics/deep learning features of 18F-FDG PET/CT image	Baseline

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Efficiency of the segmentation model	The effectiveness of the segmentation model is evaluated by the detection rate of lesions and the Dice similarity coefficient (2(A∩B)/ (A+B), A=segmented voxel volume, B=ground truth volume), which both describes the accuracy of dividing the lesion and the background.	immediately after the development and testing of models
Efficiency of the classification model	The classification model is evaluated by the accuracy [ (TP+TN)/(TP+FP+TN+FN) ] , precision [TP/(TP+FP)], recall [TP/(TP+FN)], F1-score [2*precision*recall/(precision+recall)], which all describes the ratio of correctly or wrongly classified lesions of the samples from different aspects. While the receiver operating characteristic (ROC) curve can illustrate this more visuelly. Area under the curve (AUC) calculated the proportion of area under the ROC curve, ranging from 0 to 1, representing the overall efficiency of classification in each group.	immediately after the development and testing of models

Collaborators and Investigators

• Sponsor: Ruijin Hospital
• Collaborators: Jiangsu Province Hospital of Traditional Chinese Medicine; Shanghai Pulmonary Hospital, Shanghai, China; Luan people's hospital; Ruijin North Hospital

Study record dates

Study Major Dates

• Study Start (Actual): April 1, 2024
• Primary Completion (Actual): July 20, 2024
• Study Completion (Actual): April 30, 2025

Study Registration Dates

• First Submitted: September 11, 2024
• First Submitted That Met QC Criteria: September 17, 2024
• First Posted (Actual): September 19, 2024

Study Record Updates

• Last Update Posted (Actual): July 23, 2025
• Last Update Submitted That Met QC Criteria: July 22, 2025
• Last Verified: July 1, 2025

More Information

Terms related to this study

• Keywords: lung cancer; deep learning; radiomics; positron emission tomography/computed tomography; pulmonary lymphoma
• Additional Relevant MeSH Terms: Neoplasms; Immune System Diseases; Neoplasms by Histologic Type; Lymphatic Diseases; Lymphoproliferative Disorders; Immunoproliferative Disorders; Lymphoma
• Other Study ID Numbers: RuijinH 2024-70

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No