Two sisters opened a restaurant together and were diagnosed with lung cancer at the same time! Just because of this habit that many people do...

Recently, a heartbreaking news became a hot topic: a pair of sisters who run a small restaurant together were diagnosed with lung cancer at the same time! However, the sisters are between 40 and 50 years old and have no history of smoking. Such "family clustering" of lung cancer is very rare.

The causes of lung cancer are complex and varied, and its early symptoms are not obvious, which makes diagnosis difficult. At the same time, both genetic and environmental factors may induce lung cancer. The "Chinese Medical Association Guidelines for Clinical Diagnosis and Treatment of Lung Cancer (2022 Edition)" points out that kitchen fumes produced by cooking methods such as frying can cause DNA damage or cancer, and is one of the important risk factors for lung cancer in non-smoking women in China.

Experts have inferred that the two hardworking sisters worked intensively in the kitchen all day, and the fumes contained a large amount of carcinogens such as benzopyrene, polycyclic aromatic hydrocarbons, heterocyclic amines, butadiene, etc., which is very likely the culprit of their lung cancer!

Early diagnosis of lung cancer is crucial

Fortunately, the sisters received timely physical examinations and follow-up, and the lung tumor was removed as soon as possible, minimizing the recurrence rate and mortality rate.

Lung cancer is the second most common cancer with the highest mortality rate worldwide, accounting for 21% of all cancer-related deaths. Currently, early diagnosis of lung cancer remains a serious challenge. Compared with other cancers, lung cancer has a high mortality rate. However, among confirmed cases, only about 20% are diagnosed as stage I.

What is alarming is that in addition to male patients, there are more and more young and middle-aged female patients in recent years. There are no obvious symptoms in the early stages. When symptoms such as chest pain and hemoptysis appear, they go for examination. By then, more than 70% of the patients have developed to the middle and late stages. Many people have lost their precious lives due to the serious illness.

In addition, with the rapid progress of medicine, immunotherapy and targeted therapy for lung cancer patients have made significant progress, but their efficacy is still unstable. Therefore, the development of highly sensitive and specific early diagnosis tools for lung cancer is crucial to safeguarding human health!

Artificial intelligence helps diagnose early lung cancer

In recent years, artificial intelligence (AI) has become a hot topic, attracting great interest from scientists and doctors in its potential role in lung cancer, and is expected to become a powerful tool for early diagnosis of lung cancer!

AI plays a vital role in the screening, diagnosis and treatment of lung cancer. It helps doctors analyze and interpret complex medical information, and ultimately helps doctors better diagnose, manage treatment and predict the prognosis of various clinical cases. Currently, the FDA has approved the application of artificial intelligence in multiple clinical treatment areas.

Taking lung cancer heterogeneity as an example, it has become a major area of ​​artificial intelligence application. AI mainly focuses on the modalities of lung cancer screening: imaging (nodule detection, segmentation and characterization) and non-imaging technologies.

LDCT is the gold standard for lung cancer screening and is the only method that can reduce the mortality rate of lung cancer patients. Currently, advanced prediction models are being developed that combine CT images with innovative technologies such as artificial intelligence algorithms to improve diagnostic accuracy and create a pair of "X-ray eyes" for doctors.

Interestingly, a large amount of data showed that AI diagnostic results were comparable to those of experienced radiologists. At the same time, AI combined with CXR and LDCT scans had higher accuracy in detecting lung nodules.

For example, the nodule detection rate of CT examinations based on the DL-CADe system is higher than the diagnostic accuracy of two radiologists. When faced with lung nodules, doctors usually use the size, volume and density of the nodules to determine whether they are benign or malignant. Although LDCT scans are the most commonly used method, AI can accurately measure these variables and track the growth of lung nodules during follow-up, with an accuracy rate of more than 80% for nodule classification.

In addition to the identification of lung nodules, accurate staging of lung cancer helps to develop the most appropriate treatment strategy and prognosis for patients.

AI can help accelerate the accurate staging of lung cancer and minimize the time-consuming tasks of doctors reading pathology slides and CT scans. Using AI as a second reader for PET and CT readings and predicting the anatomical location of metastatic lesions through multi-planar reconstruction of PET and CT scans not only effectively reduces the workload of radiologists, but also improves the accuracy of nodule detection. It is a win-win situation!

In addition, AI has shown the potential to combine continuous imaging data to track tumor changes over time. By leveraging DL methods and recurrent neural networks (RNNs), AI can analyze longitudinal data from CT scans of tumor patients after treatment and provide valuable insights into phenotypic characteristics and treatment responses!

Artificial intelligence has great potential in helping cancer treatment

With the development and application of algorithms such as machine learning and convolutional neural networks, the development of imaging AI has entered a new stage of explosive growth.

Taking lung cancer as an example, by combining AI models, clinical data and imaging results, doctors can be further guided to understand the patient's clinical results, develop personalized and efficient treatment plans, and improve the patient's life and health experience. This is an attractive challenge and competition.

It is worth noting that although AI has great potential in lung cancer, we have to face the challenges and limitations of its implementation. How to address data quality, improve model interpretability and ethical considerations is crucial to ensure the successful integration of AI into clinical practice!

References:

[1] Gandhi Z, Gurram P, Amgai B, Lekkala SP, Lokhandwala A, Manne S, Mohammed A, Koshiya H, Dewaswala N, Desai R, Bhopalwala H, Ganti S, Surani S. Artificial Intelligence and Lung Cancer: Impact on Improving Patient Outcomes. Cancers (Basel). 2023 Oct 31;15(21):5236. doi: 10.3390/cancers15215236. PMID: 37958411; PMCID: PMC10650618.