Background Accurate and robust pathological image analysis for colorectal cancer (CRC) diagnosis is time-consuming and knowledge-intensive, but is essential for CRC patients’ treatment. The current heavy workload of pathologists in clinics/hospitals may easily lead to unconscious misdiagnosis of CRC based on daily image analyses. Methods Based on a state-of-the-art transfer-learned deep convolutional neural network in artificial intelligence (AI), we proposed a novel patch aggregation strategy for clinic CRC diagnosis using weakly labeled pathological whole-slide image (WSI) patches. This approach was trained and validated using an unprecedented and enormously large number of 170,099 patches, > 14,680 WSIs, from > 9631 subjects that covered diverse and representative clinical cases from multi-independent-sources across China, the USA, and Germany. Results Our innovative AI tool consistently and nearly perfectly agreed with (average Kappa statistic 0.896) and even often better than most of the experienced expert pathologists when tested in diagnosing CRC WSIs from multicenters. The average area under the receiver operating characteristics curve (AUC) of AI was greater than that of the pathologists (0.988 vs 0.970) and achieved the best performance among the application of other AI methods to CRC diagnosis. Our AI-generated heatmap highlights the image regions of cancer tissue/cells. Conclusions This first-ever generalizable AI system can handle large amounts of WSIs consistently and robustly without potential bias due to fatigue commonly experienced by clinical pathologists. It will drastically alleviate the heavy clinical burden of daily pathology diagnosis and improve the treatment for CRC patients. This tool is generalizable to other cancer diagnosis based on image recognition.

No abstract

No abstract

Aims: Pancreatic cancer (PC) is a malignant tumor with a strong invasive nature and low survival rate. We aimed to estimate the PC burden at the global, regional, and national levels in 204 countries from 1990 to 2019. Methods: Detailed data, including the incidence, death, and disability-adjusted life years (DALYs), were analyzed from the Global Burden of Diseases Study 2019. Results: Globally, there were 530,297 (486,175–573,635) incident cases and 531,107 (491,948–566,537) deaths from PC in 2019. The age-standardized incidence rate (ASIR) was 6.6 (6–7.1), and the age-standardized mortality rate (ASMR) was 6.6 (6.1–7.1) per 100,000 person-years. PC caused 11,549,016 (10,777,405–12,338,912) DALYs, with an age-standardized rate of 139.6 (130.2–149.1) per 100,000 person-years. There were increases in estimated annual percentage changes (EAPCs) of ASIR (0.83; 0.78–0.87), ASMR (0.77; 0.73–0.81), and age-standardized DALYs rate (ASDR) (0.67; 0.63–0.71). The global number of incident cases increased by 168.7%, from 197,348 (188,604–203,971) to 530,297 (486,175–573,635); the number of deaths increased by 168.2% from 198,051 (189,329–204,763) to 531,107 (491,948–566,537); and total DALYs increased by 148.5% from 4,647,207 (4,465,440–4,812,129) to 11,549,016 (10,777,405–12,338,912). East Asia and China recorded the highest number of incident cases, deaths, and DALYs. The proportion of deaths was attributable to smoking (21.4%), elevated fasting glucose (9.1%), and high BMI (6%). Conclusions: Our study updated the epidemiological trends and risk factors for PC. PC remains a major hazard to the sustainability of health systems worldwide, with an increasing incidence rate and mortality from 1990 to 2019. More targeted strategies are required to prevent and treat PC.

No abstract