Endocrinology

Published by Daniel at January 1, 2022

New Thyroid Assessment Tool

In Boston, Massachusetts, Jeffrey R. Garber, MD, chief of endocrinology at Atrius Health and associate professor of medicine at Harvard Medical School, designed a new interactive tool. Thyroid Nodule App (TNAPP) is a computer-interpretable guideline that incorporates data from different update guidelines recommendations on thyroid nodule evaluation. The American Association of Clinical Endocrinology announced that the app offers a novel and comprehensive approach to improving the clinical decision-making process for thyroid nodules and appears to be a more involved calculator that tries to address several stages of decision making for thyroid nodule diagnosis and management.(1)

The American Association of Clinical Endocrinology announced that the app offers a novel and comprehensive approach to improving the clinical decision-making process for thyroid nodules and appears to be a more involved calculator that tries to address several stages of decision making for thyroid nodule diagnosis and management.(1)

Fully Automated Artificial Pancreas System for management of Type 1 Diabetes

Type 1 Diabetes is a chronic condition that requires lifelong treatment with insulin. To help accurately treat this disease, a Korean research team, led by Professor Sung-Min Park from Pohang University of Science and Technology, developed an AI algorithm that calculates the amount of insulin needed for a diabetic patient and injects it automatically. This algorithm is widely known as AlphaGo, which uses pharmacological concepts to reinforce learning and prevent errors. It helps maintain a mean glucose level of 124.72 mg/dL irrespective of the input of meal intake and exercise time.(2)

This algorithm is widely known as AlphaGo, which uses pharmacological concepts to reinforce learning and prevent errors. It helps maintain a mean glucose level of 124.72 mg/dL irrespective of the input of meal intake and exercise time.(2)

AI could detect the risk of pancreatic cancer two years before doctors

Researchers from Barts Cancer Institute at Queen Mary University of London and Edge Hill University are investigating how artificial intelligence could improve the early diagnosis of pancreatic cancer. They created a logistic regression with different models using data from electronic health records to look for the subtle signs risk groups who are more likely to develop pancreatic cancer; in combination with AI to identify potential biomarkers as carbohydrate antigen 19-9 assay. Early results showed that pancreatic cancer diagnosis could be accurately predicted for 60% of the patients younger than 60, with a sensitivity of 76% and a specificity of 45%. These findings mean that using AI to identify people at very high risk of pancreatic cancer up to 24 months earlier could make the difference between life and death.(3)

Early results showed that pancreatic cancer diagnosis could be accurately predicted for 60% of the patients younger than 60, with a sensitivity of 76% and a specificity of 45%. These findings mean that using AI to identify people at very high risk of pancreatic cancer up to 24 months earlier could make the difference between life and death.(3)

When Detecting Broken Bones, Computers Outperform People

According to the National Osteoporosis Foundation, only 2 in 10 women in the United States who suffer a break a bone are screened or treated for osteoporosis. With this information, Christopher White, of Prince of Wales Hospital in Randwick, Australia, tested a new AI-driven search tool, X-Ray Artificial Intelligence Tool (XRAIT), which can find around five times the number of broken bones on X-Ray or CT scans than a radiologist can discover by reading the reports. XRAIT’s efficacy was tested in searches of 5,089 digital radiology reports from patients over age 50 who visited the emergency room for bone imaging within three months. A comparison with radiologist performance revealed XRAIT pinpointed three times the fractures, identifying 349 people with breaks likely due to low-bone mass versus the 98 found by radiologists.(4)

XRAIT’s efficacy was tested in searches of 5,089 digital radiology reports from patients over age 50 who visited the emergency room for bone imaging within three months. A comparison with radiologist performance revealed XRAIT pinpointed three times the fractures, identifying 349 people with breaks likely due to low-bone mass versus the 98 found by radiologists.(4)

Thyroid cancer now diagnosed with machine learning-powered photoacoustic/ultrasound imaging

Currently, thyroid nodule diagnosis is performed by fine-needle aspiration biopsy (FNAB) using an ultrasound. This technique can lead to repetitive and unnecessary biopsies as almost 20% of results are inaccurate. To overcome this, a joint research team at the Pohang University of Science and Technology in Korea has proposed a new non-invasive technique by combining photoacoustic (PA) and ultrasound image technology with artificial intelligence to distinguish thyroid nodules from cancer. This is based on the fact that the oxygen saturation of malignant nodules is lower than in benign nodules. It works based on the principle that when light (laser) is irradiated on the patient’s thyroid nodule, an ultrasound signal called a PA signal is generated from the thyroid gland and the nodule. This technique will significantly reduce the number of invasive biopsies and can also be implemented for the diagnosis of cancer of other organs.(5)

This is based on the fact that the oxygen saturation of malignant nodules is lower than in benign nodules. It works based on the principle that when light (laser) is irradiated on the patient’s thyroid nodule, an ultrasound signal called a PA signal is generated from the thyroid gland and the nodule. This technique will significantly reduce the number of invasive biopsies and can also be implemented for the diagnosis of cancer of other organs.(5)

Will there be Virtual Endocrinologists in the future of medicine?

Henrik Berggren, Steady Health’s founder, proposes the replacement of conventional endocrinologists with a virtual endocrinologist who can interact through text message and video chat rather than provide in-person care. Steady Health provides the same endocrinology services as traditional clinic prescriptions, laboratory tests, and referrals, but all via a mobile app. The virtual services will set up weekly tests with the patients’ primary care provider and virtual check-ups on a regular basis. The Steady Health team strives to provide “continuous coaching” between these planned appointments. In a way, the Steady Health model turns that traditional care model on its head by de-emphasizing the importance of regular check-ups and paying more attention to what happens in between those check-ups.(6)

The Steady Health team strives to provide “continuous coaching” between these planned appointments. In a way, the Steady Health model turns that traditional care model on its head by de-emphasizing the importance of regular check-ups and paying more attention to what happens in between those check-ups.(6)

Artificial Intelligence technology company receives FDA approval to revolutionize Thyroid Ultrasounds

Medo, a technology start-up company headquartered in Singapore and Canada, has developed a device named Medo-Thyroid, which works by obtaining quick video ‘sweeps’ of each side of the thyroid gland. The AI automatically reviews this data and selects the optimal images for analysis, calculates standard lobe and isthmus measurements, and finally assists in locating, measuring, and characterizing any nodules present. Medo-Thyroid creates an individualized, interactive report with this information using the well-known TI-RADS system. This technology receives approval from FDA as the world’s first tool using Artificial Intelligence to simplify the thyroid ultrasound scanning workflow.(7)

Medo-Thyroid creates an individualized, interactive report with this information using the well-known TI-RADS system. This technology receives approval from FDA as the world’s first tool using Artificial Intelligence to simplify the thyroid ultrasound scanning workflow.(7)

Artificial Intelligence for Detecting Diabetic Eye Disease

The most common microvascular complication of diabetic disease is the development of retinopathy and macular edema, which is the leading cause of blindness in the population globally. In order to prevent permanent vision loss, early recognition for the treatment of vision-threatening complications is paramount. In 2021 the US Food and Drug Administration (FDA) granted marketing approval to IDx-DR, designed by Digital Diagnostics. This is the first artificial intelligence-based medical device to detect referable retinopathy from color fundus photographs obtained from a nonmydriatic fundus camera. IDx-DR’s approval was based on a clinical study that assessed the software’s performance on retinal images from patients with diabetes at ten different primary care sites. The platform’s sensitivity and specificity were 87% and 90%, respectively.(8)

This is the first artificial intelligence-based medical device to detect referable retinopathy from color fundus photographs obtained from a nonmydriatic fundus camera. IDx-DR’s approval was based on a clinical study that assessed the software’s performance on retinal images from patients with diabetes at ten different primary care sites. The platform’s sensitivity and specificity is 87% and 90%, respectively.(8)

References