Exploring the use of Artificial Intelligence (AI) to develop a triaging solution for Tuberculosis using chest ultrasound videos
试验速览
- 阶段
- 不适用
- 状态
- 尚未招募
- 发起方
- 入组人数
- 500
- 试验地点
- 5
- 主要终点
- Creating a dataset of Chest X-rays, HRCT and Chest Ultrasound Scans (CUS) of adult subjects.
研究概览
简要总结
The study aims to create a dataset of chest X-rays, HRCT and ultrasound scans of adult subjects as per the enrollment criteria which has 4 categories of patients with specific targets for each category namely 1. Microbiologically positive pulmonary TB patients (X ray & CBNAAT both positive). 2.Clinically positive cases of pulmonary TB (X ray positive but CBNAAT negative) 3. Non TB Patients with some chest pathology (X ray & CBNAAT negative for TB) 4. Normal Healthy Individuals (X ray & CBNAAT negative for TB and no other pathology). Data will be collected from 5 study sites (Hospitals) across 3 districts in Gujarat namely Ahmedabad, Surat & Dahod. Study Sites have been selected such that they represent the urban, rural & tribal pockets of Gujarat with the highest notifying TB hospitals in these areas. Data of the Chest USG’s will be further annotated by expert radiologist’s to identify features like consolidation, subpleural nodules etc. All this data will be used to develop an AI Algorithm for developing a screening tool that will enable frontline health workers (i.e., TBHV) take a short ultrasound video of the chest of an individual to screen for features of pulmonary tuberculosis, thereby enabling timely triaging for better diagnostics & treatment. The ultimate aim of the study is to develop AI-powered screening/triaging tool for early diagnosis of pulmonary tuberculosis.
研究设计
- 研究类型
- Observational
入排标准
- 年龄范围
- 18.00 Year(s) 至 80.00 Year(s)（—）
- 性别
- All
入选标准
- •All subjects consenting to be a part of the study. Individual consent (signed and dated informed consent form)
- •18 years and above of age 3A. For category of TB subjects, include those satisfying the following criteria – (a) Patients should have had any of the following (one or more) symptoms of pulmonary tuberculosis as identified by the clinician at the time of presentation.
- •Persistent cough for 2 weeks or more; Night sweats; Chest pain; Weight loss(unintentional); Shortness of breath; Feeling tired or weak; Fever-Body temperature of more than 100.4 degrees Fahrenheit (CDC) (b)Patients X-Ray chest showing findings suggestive of tuberculosis.(c)Additionally, patients should be Microbiologically confirmed (sputum microscopy/ CBNAAT/ TruNat) OR Clinically diagnosed TB Cases (d) Clinically stable individuals.
- •Individuals that do not require emergency medical attention. 3B. For category of Non-TB subjects, include those satisfying the following criteria – (a) In the Chest symptomatic category: Subjects having symptoms of other chest conditions/pathologies. Chest x-ray findings ruling out pulmonary tuberculosis but may have other findings.(b) In Normal subjects’ category: Subjects should be clinically stable individuals with no symptoms suggestive of pulmonary conditions. Chest x-ray findings should indicate a clear chest with no lesions.
排除标准
- •Consent not given by the individual for enrolment in the study.
- •In the TB positive subjects, exclude all clinically unstable individuals.
- •For Non-TB subjects, in the Normal individual’s category, exclude all clinically ill individuals.
结局指标
主要结局
Creating a dataset of Chest X-rays, HRCT and Chest Ultrasound Scans (CUS) of adult subjects.
时间窗: 1 year
次要结局
- 1. To use this structured dataset to develop an AI-powered screening tool for triaging pulmonary tuberculosis patients from the community(2. To anonymize the dataset by removing all Personal Identifiable Information (PII) and make it publicly available for the global research community.)