结直肠T分期数据

对DICOM格式的原始腹部CT影像数据进行病灶标注，标出病灶的TMN分期，标注规则如下：
1.肿瘤侵犯程度：TNM分期标注数据中的T部分描述了肿瘤在结直肠道壁的侵犯程度，从Tx（原发肿瘤不能确定），T（无肿瘤）到T4（肿瘤侵犯到浆膜外），这些信息对于算法训练至关重要，可以帮助算法更好地识别和分类肿瘤。
2.淋巴结转移情况：结直肠TNM分期中N部分描述了淋巴结转移的情况，从N0（无淋巴结转移）到N3（大于等于7个淋巴结转移），这些信息有助于算法更准确地预测患者的预后情况和治疗方案的效果。
3.远处转移情况：结直肠TNM分期中M部分描述了是否存在远处转移，如果有，则M为1，无则为0。这些信息有助于算法更好地评估患者的病情。
以上标注数据，将用于结直肠TNM分期人工智能算法训练，有助于提高算法的准确性和可靠性，为临床医生提供更加精准和可靠的辅助诊断和治疗建议。1.数据采集：通过医疗机构获取结直肠CT数据。这些数据包括不同阶段的胃肠肿瘤，以及正常组织和其他疾病的情况。
2.数据处理：医生对采集的结直肠CT图像进行专业分析，将没有肿瘤的数据剔除。然后按照美国癌症联合委员会（American Joint Committee on Cancer, AJCC）/国际抗癌联盟（Union for Inter-national Cancer Control, UICC）结直肠癌 TNM 分期系统（2017年第八版），对结直肠CT影像数据进行分析，得到以下几个特征信息：直肠癌原发肿瘤（T）分期；区域淋巴结（N）、临床N（cN）分期；远处转移（M）分期。
3.算法规则：根据医生标注得出的影像特征信息，与结直肠癌 TNM 分期系统算法规则进行匹配，得到结直肠癌TMN分期结果。
4.数据应用：该数据用于训练结直肠TMN分期人工智能模型，以识别结直肠癌TMN分期。

Lesion annotation is performed on raw abdominal CT imaging data in DICOM format, where the TNM staging of the lesions will be marked. The annotation rules are as follows:
1. Tumor invasion extent: The T component in the TNM staging annotation data describes the degree of tumor invasion of the colorectal bowel wall, ranging from Tx (primary tumor cannot be determined), T0 (no tumor) to T4 (tumor invading beyond the serosa). This information is critical for algorithm training, as it enables the algorithm to better identify and classify tumors.
2. Lymph node metastasis status: The N component in the colorectal TNM staging system describes the status of lymph node metastasis, ranging from N0 (no lymph node metastasis) to N3 (≥7 metastatic lymph nodes). This information helps the algorithm more accurately predict patient prognosis and the efficacy of treatment plans.
3. Distant metastasis status: The M component in the colorectal TNM staging system describes the presence of distant metastasis, with M=1 if metastasis exists and M=0 if no metastasis occurs. This information assists the algorithm in better evaluating the patient's condition.

This annotated dataset will be used for training AI algorithms for colorectal TNM staging, which helps improve the accuracy and reliability of the algorithms, providing clinicians with more precise and reliable auxiliary diagnosis and treatment recommendations.
1. Data collection: Colorectal CT data is acquired from medical institutions. These data cover gastrointestinal tumors at different stages, as well as cases of normal tissues and other diseases.
2. Data processing: Professional analysis of the collected colorectal CT images is performed by physicians, and data without tumors are excluded. Subsequently, according to the 8th edition (2017) of the colorectal TNM staging system formulated by the American Joint Committee on Cancer (AJCC) and the Union for International Cancer Control (UICC), the colorectal CT imaging data are analyzed to obtain the following characteristic information: rectal cancer primary tumor (T) staging; regional lymph node (N) and clinical N (cN) staging; distant metastasis (M) staging.
3. Algorithm matching rules: The imaging feature information annotated by physicians is matched with the algorithmic rules of the colorectal TNM staging system to derive the final colorectal TNM staging results.
4. Data application: This dataset is utilized to train AI models for colorectal TNM staging, aiming to identify the TNM stage of colorectal cancer.

搜集汇总

数据集介绍

特点

结直肠T分期数据集包含521条数据，每年更新一次，用于训练结直肠TNM分期人工智能模型，以提高算法的准确性和可靠性，为临床医生提供辅助诊断和治疗建议。