A learned 3D CNN model based on a brain MRI dataset collected from clinical cases of flat brain syndrome

This trained model has been trained on a brain MRI dataset collected from clinical cases of flat brain syndrome across Japan, primarily from Showa Medical University Hospital and its affiliated medical institutions. The collection of the dataset used for training was approved by the Institutional Review Board (IRB) for Research Involving Human Subjects at Showa Medical University, and informed consent was obtained from all participants or their legal guardians. However, as the raw data cannot be made publicly available, the trained model is being released. Software/Library Version Python 3.10.18 CUDA 11.8 NumPy 1.23.5 TensorFlow 2.12.0 ''' Python code import tensorflow as tf import numpy as np # Load model Model = tf.keras.models.load_model('model_collection/ResNet18') # Show model sturacture Model.summary() # Input Data is 3D MRI data input_image = np.load('input_image_path.npy') prediction = Model(input_image.reshape((1, 64, 128, 128, 1))) print(f'prediction_class:DCX = 0, LIS1 = 1, Other = 2') print(f'prediction_result:{np.argmax(prediction)}') '''本训练模型基于日本境内扁平脑综合征临床病例采集的脑部磁共振成像（MRI）数据集训练而成，该数据集主要采集自昭和大学医院及其附属医疗机构。 本次训练所用数据集的采集工作，已通过昭和大学涉及人类受试者研究的伦理审查委员会（IRB）审批，且所有受试者或其法定监护人均已签署知情同意书。 但由于原始数据无法公开获取，故本次仅发布该训练模型。 软件/库 版本 Python 3.10.18 CUDA 11.8 NumPy 1.23.5 TensorFlow 2.12.0 ''' Python代码 import tensorflow as tf import numpy as np # 加载模型 Model = tf.keras.models.load_model('model_collection/ResNet18') # 查看模型结构 Model.summary() # 输入数据为三维MRI数据 input_image = np.load('input_image_path.npy') prediction = Model(input_image.reshape((1, 64, 128, 128, 1))) print(f'预测类别：DCX = 0, LIS1 = 1, 其他 = 2') print(f'预测结果：{np.argmax(prediction)}') '''