Background We aim to use Natural Language Processing (NLP) to automate the extraction and classification of thyroid cancer risk factors from pathology reports. Methods We analyzed 1,410 surgical pathology reports from adult papillary thyroid cancer patients at Mayo Clinic, Rochester, MN, from 2010 to 2019. Structured and non-structured reports were used to create a consensus-based ground truth dictionary and categorized them into modified recurrence risk levels. Non-structured reports were narrative, while structured reports followed standardized formats. We then developed ThyroPath, a rule-based NLP pipeline, to extract and classify thyroid cancer features into risk categories. Training involved 225 reports (150 structured, 75 unstructured), with testing on 170 reports (120 structured, 50 unstructured) for evaluation. The pipeline's performance was assessed using both strict and lenient criteria for accuracy, precision, recall, and F1-score. Results In extraction tasks, ThyroPath achieved overall strict F-1 scores of 93% for structured reports and 90 for unstructured reports, covering 18 thyroid cancer pathology features. In classification tasks, ThyroPath-extracted information demonstrated an overall accuracy of 93% in categorizing reports based on their corresponding guideline-based risk of recurrence: 76.9% for high-risk, 86.8% for intermediate risk, and 100% for both low and very low-risk cases. However, ThyroPath achieved 100% accuracy across all thyroid cancer risk categories with human-extracted pathology information. Conclusions ThyroPath shows promise in automating the extraction and risk recurrence classification of thyroid pathology reports at large scale. It offers a solution to laborious manual reviews and advancing virtual registries. However, it requires further validation before implementation.

Uterine leiomyosarcoma (LMS) is a rare but aggressive malignancy. On imaging, it is difficult to differentiate LMS from, for example, degenerated leiomyoma (LM), a prevalent but benign condition. We curated a data set of 115 axial T2-weighted MRI images from 110 patients (mean [range] age=45 [17-81] years) with UTs that included five different tumor types. These data were randomly split stratifying on tumor volume into training (n=85) and test sets (n=30). An independent second reader (reader 2) provided manual segmentations for all test set images. To automate segmentation, we applied nnU-Net and explored the effect of training set size on performance by randomly generating subsets with 25, 45, 65 and 85 training set images. We evaluated the ability of radiomic features to distinguish between types of UT individually and when combined through feature selection and machine learning. Using the entire training set the mean [95% CI] fibroid DSC was measured as 0.87 [0.59-1.00] and the agreement between the two readers was 0.89 [0.77-1.0] on the test set. When classifying degenerated LM from LMS we achieve a test set F1-score of 0.80. Classifying UTs based on radiomic features we identify classifiers achieving F1-scores of 0.53 [0.45, 0.61] and 0.80 [0.80, 0.80] on the test set for the benign versus malignant, and degenerated LM versus LMS tasks. We show that it is possible to develop an automated method for 3D segmentation of the uterus and UT that is close to human-level performance with fewer than 150 annotated images. For distinguishing UT types, while we train models that merit further investigation with additional data, reliable automatic differentiation of UTs remains a challenge.

A podcast devoted to clinical radiology and allied sciences, owned and published by the Radiological Society of North America.

U.K. software developer Ultromics has raised 28 million euros, which it plans to use to further develop artificial-intelligence (AI)-supported echocardiograms. The company was founded in 2017 as a research spin-off from the University of Oxford in England. It has developed two platforms for assessing heart function: EchoGo Core and EchoGo Pro. Both are already being used by U.S. organizations, including the Mayo Clinic in Rochester, MN, and the Oregon Health and Science University in Portland. The funding round was led by the Blue Venture Fund -- a collaboration of Blue Cross Blue Shield firms, the Blue Cross Blue Shield Association, and Sandbox -- with participation from Optum Ventures and existing investor Oxford Sciences Innovation, the company said.

FRIDAY, May 28, 2021 (HealthDay News) -- An artificial intelligence (AI) algorithm that uses data from electrocardiography can help increase the diagnosis of low ejection fraction (EF), according to a study published online May 6 in Nature Medicine. Xiaoxi Yao, Ph.D., from the Mayo Clinic in Rochester, Minnesota, and colleagues randomly assigned 120 primary care teams, including 358 clinicians, to intervention (access to AI results from the low ejection fraction algorithm developed by Mayo and licensed to Anumana Inc.; 181 clinicians) or control (usual care; 177 clinicians) in a pragmatic trial at 45 clinics and hospitals. A total of 22,641 adult patients with echocardiography performed as part of routine care were included (11,573 in the intervention group; 11,068 controls). The researchers found positive AI results, indicating a high likelihood of low EF, in 6.0 percent of patients in both arms. More echocardiograms were obtained for patients with positive results by clinicians in the intervention group (49.6 versus 38.1 percent), but echocardiogram use was similar in the overall cohort (19.2 versus 18.2 percent).

Xiaoxi Yao, Ph.D., from the Mayo Clinic in Rochester, Minnesota, and colleagues randomly assigned 120 primary care teams, including 358 clinicians, to intervention (access to AI results from the low ejection fraction algorithm developed by Mayo and licensed to Anumana Inc.; 181 clinicians) or control (usual care; 177 clinicians) in a pragmatic trial at 45 clinics and hospitals. More echocardiograms were obtained for patients with positive results by clinicians in the intervention group (49.6 versus 38.1 percent), but echocardiogram use was similar in the overall cohort (19.2 versus 18.2 percent). The diagnosis of low EF was increased with the intervention in the overall cohort (2.1 versus 1.6 percent; odds ratio, 1.32) and among patients with positive results (19.5 versus 14.5 percent; odds ratio, 1.43). "The AI intervention increased the diagnosis of low ejection fraction overall by 32 percent relative to usual care.

Major depressive disorder (MDD) is a prevalent psychiatric disorder that is associated with significant healthcare burden worldwide. Phenotyping of MDD can help early diagnosis and consequently may have significant advantages in patient management. In prior research MDD phenotypes have been extracted from structured Electronic Health Records (EHR) or using Electroencephalographic (EEG) data with traditional machine learning models to predict MDD phenotypes. However, MDD phenotypic information is also documented in free-text EHR data, such as clinical notes. While clinical notes may provide more accurate phenotyping information, natural language processing (NLP) algorithms must be developed to abstract such information. Recent advancements in NLP resulted in state-of-the-art neural language models, such as Bidirectional Encoder Representations for Transformers (BERT) model, which is a transformer-based model that can be pre-trained from a corpus of unsupervised text data and then fine-tuned on specific tasks. However, such neural language models have been underutilized in clinical NLP tasks due to the lack of large training datasets. In the literature, researchers have utilized the distant supervision paradigm to train machine learning models on clinical text classification tasks to mitigate the issue of lacking annotated training data. It is still unknown whether the paradigm is effective for neural language models. In this paper, we propose to leverage the neural language models in a distant supervision paradigm to identify MDD phenotypes from clinical notes. The experimental results indicate that our proposed approach is effective in identifying MDD phenotypes and that the Bio- Clinical BERT, a specific BERT model for clinical data, achieved the best performance in comparison with conventional machine learning models.

Google has been partnering with the Mayo Clinic, the top hospital in the United States, for years now. But today, the company announced a move to make that partnership a little more serious: Google will open an office in Rochester MN this year, home of the Mayo Clinic's headquarters. Google already has dozens of offices in the US, but this will be the first in Minneapolis. The company worked with the Mayo Clinic to find spaces that let them collaborate easily with Mayo Clinic staff as they look to "transform patient care." Announcing a new office during the COVID-19 pandemic might seem a bit unusual at first glance, given that companies and employees alike are embracing remote work more than ever before. Google acknowledges that they'll be waiting to fully open until later this year and will wait until state and local COVID-19 guidelines say it's safe to do so.

One of the effects of the COVID-19 public health emergency is that it has added urgency and speed to technology transformations that were already occurring, such as cloud migration and deployments of artificial intelligence and machine learning. At few places is that shift more pronounced than at Rochester, Minnesota-based Mayo Clinic, which six months before the pandemic arrived in the United States had embarked on a decade-long strategic partnership with Google Cloud. "Our partnership will propel a multitude of AI projects currently spearheaded by our scientists and physicians, and will provide technology tools to unlock the value of data and deliver answers at a scale much greater than today," said Mayo CIO Cris Ross at the time. Shortly after the partnership was announced, toward the end of 2019, the health system hired longtime CIO Dr. John Halamka as president of Mayo Clinic Platform, tasking him with leading a cloud-hosted, AI-powered digital transformation across the enterprise. In the months since, like the rest of the world, Mayo Clinic has found itself tested and challenged by the pandemic and its ripple effect – but has also embraced the moment as an inflection point, a powerful moment to push forward with an array of new use cases to drive quality improvement, streamline efficiency, and boost the health of patients and populations in the years ahead.

In a letter published in the New England Journal of Medicine, researchers at the Mayo Clinic showed that the required steps are not complex. But privacy experts questioned whether the process could be replicated on a much larger scale with today's technology. The subjects were 84 healthy participants in a long-term study of about 2,000 residents of Olmsted County, Minn. Participants get brain scans to look for signs of Alzheimer's disease, as well as cognitive, blood and genetic tests. Over the years, the study has accumulated over 6,000 M.R.I. scans. After the participants agreed to the experiment, a team led by Christopher Schwarz, a computer scientist at the Mayo Clinic, photographed their faces and, separately, used a computer program to reconstruct faces from the M.R.I.'s.