Risk Model for Metastasis Detection of Neuroblastoma

Overview

Neuroblastoma (NB) is the most common extracranial solid tumor in children, accounting for about 15% of tumor-related mortality. NB patients in high-risk group are prone to bone marrow and/or bone metastases with low five-year overall survival rate. The artificial intelligence (AI) and deep learning technologies have potential to identifying morphological characteristics of bone marrow cytology in clinical practice. In this study, the investigators construct and evaluate the bone marrow cytology-based AI model for detection and prognosis of NB. The main questions of the study as follows: The question 1: Dose bone marrow cytology-based AI model work for prediction of bone marrow metastasis in NB? The question 2: Dose bone marrow cytology-based AI model work for prediction of bone metastasis in NB? The question 3: Dose bone marrow cytology-based AI model have potential to assist doctors in making individualized predictions of survival outcome? The investigators will retrospectively obtain the participants with NB between January 2019 and June 2024. The follow-up date ended on June 30, 2024. The internal cohort including participants from Xinhua Hospital, Shanghai Jiao Tong University School of Medicine. The independent external cohorts including participants form Children's Hospital, Zhejiang University School of Medicine and Shenzhen Children's Hospital. The investigators collect the clinical data of enrolled participants at the time of the patients' initial admission to the hospital, prior to receiving treatment. The clinical information including age, gender, primary tumor location, tumor grade, bone marrow metastasis state, bone metastasis state, genetic aberrations (MYCN amplification, Chromosome 1p deletion, Chromosome 11q deletion) and lab variables (peripheral blood cell count, bone marrow cytology indicators, the serum concentration of lactate dehydrogenase, neuron specific enolase). This study is a non-interventional observational study, there is no risk to the participants and investigators. Participants get these benefits: 1. Early Detection: The model helps in early risk identification and personalize treatment. 2. Convenience: Because the model relies on general lab tests, it is easy to carry out can reduce invasive diagnostic procedures. 3. Cost-Effective: Using existing clinical data from routine tests can make the prediction process more cost-effective. 4. Data-Driven Decisions: The AI model improve diagnostic efficiency and support the medical decision.

1. Dataset processing. In this study, the model including training set, validation set and test sets. During the model construction, the investigators process the training in internal cohort by randomly assign the enrolled participants to the training set and validation set in a 8:2 ratio. The investigators use the enrolled participants of external cohorts as test sets to evaluate the model performance independently, and then select the best model for future use. The Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiaotong University School of Medicine apaproved this study (XHEC-C-2024-023-2). 2. image acquisition. In the AI model, the investigators use bone marrow smears of enrolled participants for cytological evaluation and image collection. During cell classification and metastasis detection, the experienced pathologists complete the bone marrow smear analysis, metastatic NB clusters in bone marrow usually exhibited aggregated round atypical cells with high nucleus/cytoplasmic ratio. The investigators scan stained bone marrow smear at 40 × magnification for digital whole slide imaging (WSI). The investigators segment WSIs into smaller patches as 512 × 512 pixels tiles and apply the Vahadane method to normalize the color of small tiles. 3. Deep learning training. In the feature extraction of bone marrow cytological image, the process including two tiers of predictions: patch-level and WSI-level predictions. For patch-level predictions, the investigators carry out label predictions and their respective probabilities for all patches. The investigators apply model in the deep learning process as follows: recognized neural network (CNN)-resnet50 and Vision Transformer (ViT). The parameter configurations in the model as follows: optimizer-SGD, loss function-softmax cross-entropy, with a batch size of 64. For WSI-level predictions, the investigators use a multi-instance learning (MIL) algorithm to aggregate dispersed patch-level features to WSI-level features. During MIL for WSI fusion, the investigators perform WSI-level predictions with Patch Likelihood Histogram (PLH) pipeline and Bag of Words (BoW) pipeline in combination. Subsequently, the investigators get the WSI-level prediction as final representations of participant for subsequent analytical operations. 4. Signature building. In the procession of feature selection, the investigators use LASSO (Least Absolute Shrinkage and Selection Operator) feature screening to determine the final WSI-level features of the bone marrow cytology. These selected features were then subjected to machine learning methods to develop AI model. the investigators apply several machine learning algorithms to predict metastasis state of bone marrow and/or bone in participants, such as support vector machines (SVM), Logistic regression (LR), tree-based models, such as random forests and extremely randomized trees (ExtraTrees), extreme Gradient Boosting (XGBoost), and light gradient boosting machine (LightGBM), as well as multilayer perceptron (MLP) to develop our models. In the prognosis model, the investigators e use Cox models to construct the survival model with bone marrow cytological signature and clinical characters. 5. Model evaluation and statistical analysis. The investigators compare the clinical characteristics of participant with independent sample t-test for continuous variables and the χ² test for discrete variables in SPSS version 22.0 (SPSS, Inc., Chicago, IL, USA). P ≤ 0.05 was considered statistically significant. For the diagnostic model, the investigators use both micro and macro area under the curve (AUC) metrics to evaluate the model in terms of sensitivity, specificity, accuracy, positive predictive value and negative predictive value at different classification thresholds. The model's performance and effectiveness were evaluated on separate test cohort. The study employs custom Python code written in Python v.3.7.12 to evaluate the model performance. For the prognostic model, the investigators use AUC as the performance metric and calculating sensitivity and specificity. The model with the best performance on the test set was selected as the optimal model. Survival curves were constructed according to the Kaplan-Meier method.