Riverain公司的人工智能技术有助于早期肺癌的诊断

第一种用于胸部X线早期肺癌检测的人工智能系统获得了美国食品药品管理局的许可。在临床试验中，放射科医生使用RapidScreen计算机辅助检测技术，对早期肺癌的检测率增加了23％，肺部肿块检测范围在9～14毫米之间。该计算机辅助检测系统是由Riverain医疗公司(Dayton OH, USA)研制成功的。

如果肺癌早期得到发现与治疗，每年仅仅在美国就可以拯救54,000例患者的生命。当前，只有15％的肺癌患者能够在肿瘤实体大小为9～14毫米的时候发现。在这个范围的肺部小肿瘤，如果得到治疗将获得令人满意的效果。在这个阶段发现的肺癌患者据估计有85%能够得到控制或者治愈。

“使用RapidScreen作为常规胸部X线初步筛查方法，补充的检查方法，比如CT(计算机X线断层扫描技术）可以用于进一步鉴别与证实肺部结节的性质，”美国乔治敦大学医学中心(Washington DC, USA)的Matthew Freedman医生指出。Freedman医生是研制用于早期肺癌诊断计算机辅助检测技术的先驱者。“胸部X线计算机辅助检测提高了胸部放射检查的价值。临床试验已经证实计算机辅助检测技术能够增强放射科医生在常规X线检查中发现可疑肺部结节的能力。”

AI Technology Diagnoses Early Lung Cancer

The first artificial intelligence (AI) system for chest x-ray early-stage lung cancer detection has been cleared by the U.S. Food and Drug Administration (FDA). In clinical trials, radiologists working with the RapidScreen computer-aided detection (CAD) technology demonstrated a 23% increase in detection accuracy for lung modules ranging from 9-14 mm in size. The CAD system was developed by Riverain Medical (Dayton OH, USA).

Detecting and treating more lung cancers in the earliest stage has the potential to save 54,000 lives annually in the United States alone. Currently, only 15% of lung tumors are found when they are small solitary pulmonary nodules in the 9-14 mm size. Small lung tumors detected within this range present the most significant prospect for earlier treatment for patients. It is estimated that lung tumors found at this stage may be controlled or cured in up to 85% of all patients.

“After initial screening and detection with RapidScreen using routine chest x-rays, complementary technologies such as CT [computed topography] may be used to further characterize and confirm nodule identity,” stated Matthew Freedman, M.D., from Georgetown University Medical Center (Washington DC, USA). Dr. Freedman is a pioneer in CAD technology for early-stage lung cancer diagnosis. “Chest x-ray CAD enhances the value of the chest radiograph. Clinical trials have demonstrated that CAD prompt marks can increase the ability of radiologists to find suspicious nodules in routine x-rays.”