Detection of Metastatic Tumor Cells in the Bone Marrow Aspirate Smears by Artificial Intelligence (AI)-Based Morphogo System

IntroductionMetastatic carcinomas of bone marrow (MCBM) are characterized as tumors of non-hematopoietic origin spreading to the bone marrow through blood or lymphatic circulation. The diagnosis is critical for tumor staging, treatment selection and prognostic risk stratification. However, the identification of metastatic carcinoma cells on bone marrow aspiration smears is technically challenging by conventional microscopic screening.ObjectiveThe aim of this study is to develop an automatic recognition system using deep learning algorithms applied to bone marrow cells image analysis. The system takes advantage of an artificial intelligence (AI)-based method in recognizing metastatic atypical cancer clusters and promoting rapid diagnosis.MethodsWe retrospectively reviewed metastatic non-hematopoietic malignancies in bone marrow aspirate smears collected from 60 cases of patients admitted to Zhongshan Hospital. High resolution digital bone marrow aspirate smear images were generated and automatically analyzed by Morphogo AI based system. Morphogo system was trained and validated using 20748 cell cluster images from randomly selected 50 MCBM patients. 5469 pre-classified cell cluster images from the remaining 10 MCBM patients were used to test the recognition performance between Morphogo and experienced pathologists.ResultsMorphogo exhibited a sensitivity of 56.6%, a specificity of 91.3%, and an accuracy of 82.2% in the recognition of metastatic cancer cells. Morphogo’s classification result was in general agreement with the conventional standard in the diagnosis of metastatic cancer clusters, with a Kappa value of 0.513. The test results between Morphogo and pathologists H1, H2 and H3 agreement demonstrated a reliability coefficient of 0.827. The area under the curve (AUC) for Morphogo to diagnose the cancer cell clusters was 0.865.ConclusionIn patients with clinical history of cancer, the Morphogo system was validated as a useful screening tool in the identification of metastatic cancer cells in the bone marrow aspirate smears. It has potential clinical application in the diagnostic assessment of metastatic cancers for staging and in screening MCBM during morphology examination when the symptoms of the primary site are indolent.

Cancer and industrial activities in China

Associations between pollution and life expectancy, infant mortality, and cardiorespiratory disease are documented in China. Yet, less is known about environmental drivers of Chinese cancers. Here, we systematically link polluting industrial activity to cancer incidence, cancer mortality, and cancer cluster designations. We investigate county-level associations between industrial production and age-adjusted incidence and mortality reported in official cancer registries. We then combine the locations of roughly 3 million enterprises with administrative data from roughly 600,000 villages and cancer cluster documentation from 380 villages. We show that county-level value-added from industry is associated with age-adjusted incidence and mortality for all cancers; bronchus, trachea, and lung cancers; stomach cancers; and esophageal cancers. We show that the odds that a village contains a documented cancer cluster increase 3-4 times if the village contains a pollution-intensive industrial facility. Leather, chemical, and dye enterprises appear to drive results. All else equal, smaller facilities increase the odds of cancer clusters.

Applying Social Science to Bring Resident Stakeholders into Pollution Governance: A Rural Environmental Justice Public Health Case Study

The purpose of this engaged public sociology study was to use social science to bring resident stakeholders into the process of governing pollution production in a rural community. The community has cancer clusters. Residents have concerns about direct exposure to pollution production in their neighborhood by a steel recycling plant that has been cited numerous times for environmental violations. The facility has been under voluntary remediation since 2009, but neighborhood residents were marginalized from the governance process. This case study details how social science was used to bring neighborhood residents’ concerns about direct exposure to toxic air pollution into remediation governance. A curricula-as-research model was developed to provide an engagement framework that guided the case study as it progressed through a series of six stages over five years. The principal investigator maintained this collaboration by integrating the project into courses, securing small grants, developing an affordable air pollution monitoring method, and convening multiple community meetings. The air monitoring results are analyzed and discussed. Finally, the impact of the case study on the company, the state environmental management agency, local government, the nonprofit partner, and residents’ sense of human agency is evaluated.

Worm-Based Microfluidic Biosensor for Real-Time Assessment of the Metastatic Status

Background: Metastasis is a complex process that affects patient treatment and survival. To routinely monitor cancer plasticity and guide treatment strategies, it is highly desired to provide information about metastatic status in real-time. Here, we proposed a worm-based (WB) microfluidic biosensor to rapidly monitor biochemical cues related to metastasis in a well-defined environment. Compared to conventional biomarker-based methods, the WB biosensor allowed high throughput screening under low cost, requiring only visual quantification of outputs; Methods: Caenorhabditis elegans were placed in the WB biosensor and exposed to samples conditioned with cancer cell clusters. The chemotactic preference of these worms was observed under discontinuous imaging to minimize the impact on physiological activity; Results: A chemotaxis index (CI) was defined to standardize the quantitative assessment from the WB biosensor, where moderate (3.24–6.5) and high (>6.5) CI levels reflected increased metastasis risk and presence of metastasis, respectively. We demonstrated that the secreted metabolite glutamate was a chemorepellent, and larger clusters associated with increased metastatic potential also enhanced CI levels; Conclusions: Overall, this study provided a proof of concept for the WB biosensors in assessing metastasis status, with the potential to evaluate patient-derived cancer clusters for routine management.

Cancer cluster and citizen alarms: epidemiological and statistical approaches

Background Respond to alarms for possible cancer cluster is a public health problem, but the management of these alarms is difficult and sometime conflictual. Methods We reviewed the guidelines on the management of disease clusters developed in various countries and the approaches used in previous disease cluster episodes in Tuscany. Statistical approaches for cluster detection were also reviewed. We performed a clustering analysis (spatio or spatio-temporal when appropriate) on childhood leukemia using data from Tuscany Cancer Registry (RTT). We used spatial hierarchical Bayesian model on aggregate data at municipality level. We performed test for general clustering and cluster detection on individual data. Results More than 100 tests for clustering analysis has been identified in the literature. Bayesian analysis on aggregate data did not show areas at higher risk with the exception of the city of Florence for childhood cancer among males. We did not found clusters for leukemia. In previous studies on disease clusters in Tuscany, most investigations have been started from community concerns and in the majority of situations a multidisciplinary approach was used. In some case an increase of incidence rate was observed, but rarely specific cluster cancer tests were used. Conclusions Hierarchical Bayesian models to aggregate data provided useful to identify long range geographical patterns while clustering analysis on individual data is a useful tool for small scale patterns. Both represent important tools for epidemiological surveillance studies particularly on childhood cancer. The best test for all situations doesn't exist, but the choice is determined by the type of question being asked from the data, by different situations and by different approaches. The Tuscany cancer clusters survey and the review of the guidelines on the management of clusters developed in different countries, give us the opportunity to formulate some suggestions for the health agencies. Key messages Respond to alarms for cluster of cancer and suggest recommendations for epidemiological and statistical standardized approaches is a public health issue. Tuscany cancer clusters survey and the review of the guidelines on the management of clusters developed in different countries, give the opportunity to formulate some suggestions for health agencies.