Application of A.I in Neuro-Oncology
October 1, 2021
1837-1900
October 1, 2021Cancer Management
Deep learning helps doctors predict gastric cancer metastasis
To help pathologists identify metastatic lymph nodes in gastric cancer, researchers at Xidian University and Changhai Hospital in China have created a computational model to predict the clinical outcomes of patients with gastric cancer. The multidisciplinary group developed a deep-learning framework that recognizes and analyzes lymph nodes’ micrometastases. It calculates the tumor area to metastatic lymph node ratio (T/MLN) from whole slide images. This AI-assisted analysis translates into better prognoses by enhancing pathologists’ performance, increasing early detection, and cutting delays in therapy initiation.(1)
To help pathologists identify metastatic lymph nodes in gastric cancer, researchers at Xidian University and Changhai Hospital in China have created a computational model to predict the clinical outcomes of patients with gastric cancer.
The multidisciplinary group developed a deep-learning framework that recognizes and analyzes lymph nodes’ micrometastases. It calculates the tumor area to metastatic lymph node ratio (T/MLN) from whole slide images. This AI-assisted analysis translates into better prognoses by enhancing pathologists’ performance, increasing early detection, and cutting delays in therapy initiation.(1)
Reveal High Performance of AI in Breast Cancer Detection
The South Korean company Lunit has developed an algorithm that accurately identifies breast cancer and outperforms other AI systems. Its purpose is to reduce radiologists’ paperwork and workload by reliably classifying mammography screenings. Based on 8,805 cases, Lunit’s algorithm showed the best accuracy and a 15-point higher sensitivity compared to the other contendants. It also surpassed the first-reader radiologists’ sensitivity and achieved a synergistic sensitivity of 88.6% when combining first-readers and Lunit’s AI solution.(2)
The South Korean company Lunit has developed an algorithm that accurately identifies breast cancer and outperforms other AI systems. Its purpose is to reduce radiologists’ paperwork and workload by reliably classifying mammography screenings. Based on 8,805 cases, Lunit’s algorithm showed the best accuracy and a 15-point higher sensitivity compared to the other contendants. It also surpassed the first-reader radiologists’ sensitivity and achieved a synergistic sensitivity of 88.6% when combining first-readers and Lunit’s AI solution.(2)
MRI data can predict patient response to treatment and survival in advanced ovarian cancer
The study “Diffusion-weighted MRI in Advanced Epithelial Ovarian Cancer Apparent Diffusion Coefficient as a Response Marker” was funded by Cancer Research UK and published in Radiology. It found that an MRI scan may predict treatment response in patients with ovarian cancer. In the study, scientists explored the Apparent Diffusion Coefficient, a calculation that describes the spatial diffusion of water molecules within tissues over time. They observed that the apparent diffusion coefficient increases more in patients whose cancer responds to treatment than in those who do not.(3
The study “Diffusion-weighted MRI in Advanced Epithelial Ovarian Cancer Apparent Diffusion Coefficient as a Response Marker” was funded by Cancer Research UK and published in Radiology. It found that an MRI scan may predict treatment response in patients with ovarian cancer. In the study, scientists explored the Apparent Diffusion Coefficient, a calculation that describes the spatial diffusion of water molecules within tissues over time.
They observed that the apparent diffusion coefficient increases more in patients whose cancer responds to treatment than in those who do not.(3)
How AI simplifies data management for drug discovery
Calithera Biosciences, in Northern California, is an immunotherapy company. It has been running clinical trials to study the safety and effectiveness of its products in patients with specific cancer-gene mutations. They seek to identify accuracy, completeness, compliance with regulations, and other regulatory aspects. An AI system alerts when a missing test result or required diary entry is found. Moreover, it can detect and cancel ransomware attacks, secure data access to specific credentials, and document incidents and reports adjusted to FDA standards or other regulatory bodies.(4)
Calithera Biosciences, in Northern California, is an immunotherapy company. It has been running clinical trials to study the safety and effectiveness of its products in patients with specific cancer-gene mutations. They seek to identify accuracy, completeness, compliance with regulations, and other regulatory aspects.
An AI system alerts when a missing test result or required diary entry is found. Moreover, it can detect and cancel ransomware attacks, secure data access to specific credentials, and document incidents and reports adjusted to FDA standards or other regulatory bodies.(4)
Artificial intelligence will be used to develop personalized treatments
The pharmaceuticals firm GlaxoSmithKline has struck a five-year partnership with King’s College London to train AI algorithms from repetitive processes rather than human instructions. The team will use a 3D cancer cell model to study how tumor from the patient’s undergoing treatment interacts with immune cells. Moreover, the model is programmed to monitor dynamic biomarkers that can predict resistance or future relapse. The partnership nourishes a novel machine learning model that integrates multimodal data, genetic and molecular traits, tumor location, images, and biomarker blood tests. The product of this integration is a “digital biological twin” of the patient to test multiple drugs and doses at numerous time points and assess its responsiveness.(5)
The pharmaceuticals firm GlaxoSmithKline has struck a five-year partnership with King’s College London to train AI algorithms from repetitive processes rather than human instructions. The team will use a 3D cancer cell model to study how tumor from the patient’s undergoing treatment interacts with immune cells. Moreover, the model is programmed to monitor dynamic biomarkers that can predict resistance or future relapse.
The partnership nourishes a novel machine learning model that integrates multimodal data, genetic and molecular traits, tumor location, images, and biomarker blood tests. The product of this integration is a “digital biological twin” of the patient to test multiple drugs and doses at numerous time points and assess its responsiveness.(5)
Researchers develop AI application able to predict 10-year prostate cancer mortality
Researchers from the radiation oncology department at the European Hospital Georges Pompidou (HEGP) and Stanford University School of Medicine have developed a new AI prediction tool for prostate cancer patients. A novel machine learning-based approach produced a 10-year mortality predictor, using only standard clinicopathological variables. Future integration of additional data will likely improve model performance rendering precise outcomes.(6)
Researchers from the radiation oncology department at the European Hospital Georges Pompidou (HEGP) and Stanford University School of Medicine have developed a new AI prediction tool for prostate cancer patients. A novel machine learning-based approach produced a 10-year mortality predictor, using only standard clinicopathological variables. Future integration of additional data will likely improve model performance rendering precise outcomes.(6)
Applying AI To Transform Early Lung Cancer Treatment
Optellum, a lung-health AI company, has entered into a strategic collaboration with the Lung Cancer Initiative at Johnson & Johnson. Through the partnership, Optellum will apply its AI-powered clinical decision support platform for early interventions and prevention of lung cancer. An AI-powered digital biomarker, generated through neural networks and imaging processing, identifies and stratifies patients’ risk by assigning a Lung Cancer Prediction score to lung nodules. The Optellum AI will be used to accurately aid in early diagnosis and optimize treatment decisions for patients in earlier stages, increasing survival rates. Optellum’s software achieved US Food and Drug Administration (FDA) clearance in March 2021 and is being implemented in clinical care by leading hospitals across the United States, with rollouts in select Asia-Pacific and European markets to follow.(7)
Optellum, a lung-health AI company, has entered into a strategic collaboration with the Lung Cancer Initiative at Johnson & Johnson. Through the partnership, Optellum will apply its AI-powered clinical decision support platform for early interventions and prevention of lung cancer. An AI-powered digital biomarker, generated through neural networks and imaging processing, identifies and stratifies patients’ risk by assigning a Lung Cancer Prediction score to lung nodules.
The Optellum AI will be used to accurately aid in early diagnosis and optimize treatment decisions for patients in earlier stages, increasing survival rates. Optellum’s software achieved US Food and Drug Administration (FDA) clearance in March 2021 and is being implemented in clinical care by leading hospitals across the United States, with rollouts in select Asia-Pacific and European markets to follow.(7)
References