AI Spotlight: Forecasting ICU Patient States for Improved Outcomes

Girish Nadkarni, MD, MPH, and Faris Gulamali

Artificial intelligence (AI) and machine learning (ML) have seen increasing use in health care, from guiding clinicians in diagnosis to helping them decide the best course of treatment. However, AI still has much unrealized potential in various health care settings.

Mount Sinai researchers are exploring bringing AI into intensive care, and developed Spatial Resolved Temporal Networks (SpaRTeN), a model to assess high-frequency patient data and generate representations of their state in real time.

The work was presented at the Time Series Representation Learning for Health workshop on Friday, May 5, hosted by the International Conference for Learned Representations, a premier gathering dedicated to machine learning.

Hear from Girish Nadkarni, MD, MPH, Irene and Dr. Arthur Fishberg Professor of Medicine at the Icahn School of Medicine at Mount Sinai and the leader of the SpaRTeN research, and Faris Gulamali, medical student at Icahn Mount Sinai and member of the Augmented Intelligence in Medicine and Science lab, on what lay behind creating the model and what it could achieve for patients.

What was the motivation for your study?

A growing amount of research is indicating the need to redefine critical illness by biological state rather than a non-specific illness syndrome. Advances in genomics, data science, and machine learning have generated evidence of different underlying etiologies for common ICU syndromes. As a result, patients with the exact same diagnosis can have entirely different outcomes.

What are the implications?

In the ICU, representations of a patient can be used to guide personalized treatments based on personalized diagnoses rather than generic treatments with empirical diagnoses.

What are the limitations of the study?

In this study, we only looked at using one type of data at a time in real time. For example, we looked primarily at measures of intracranial pressure. However, the ICU has many types of data being output simultaneously. Future work hopes to integrate all the different types of data such as electrocardiograms, blood pressure, and imaging to improve patient representations.

How might these findings be put to use?

These patient representations are being combined with data on medications and procedures to determine how to optimize patient treatment based on underlying state rather than common illness syndromes.

What is your plan for following up on this study?

In this study, we focused primarily on creating the algorithm and showing that it works for the case of intracranial hypertension. In future studies, we would like to integrate multiple data modalities such as imaging, electrocardiograms, and blood pressure as well as intervention-based data such as medications and procedures to determine precise empirical interventions that lead to improvements in short-term and long-term patient outcomes.

Learn more about how Mount Sinai researchers and clinicians are leveraging machine learning to improve patient lives

More than 50 years after the United States formally declared war on cancer, what is the prognosis for innovative cancer research and care?

Two Mount Sinai leaders in cancer care and research, Ramon Parsons, MD, PhD, Director of The Tisch Cancer Institute at the Icahn School of Medicine at Mount Sinai, and Cardinale Smith, MD, PhD, Vice President, Cancer Clinical Services for the Mount Sinai Health System, offered their perspectives as part of a recent 92nd Street Y online event. You can watch the video here.

The two agreed on this overarching message: Tremendous progress has been made in unraveling the complex biology of cancer and targeting its many forms with advanced new medicines, particularly immunotherapies. But looming just as large are the challenges that remain in areas like overcoming resistance to these medicines, early detection of cancer through screening, and ensuring the equitable distribution of cancer care to diverse and disadvantaged populations.

Ramon Parsons, MD, PhD

“After 50 years we have a much more sophisticated understanding of how genes are altered in the cancer cell and how cancer cells reprogram the tumor microenvironment,” said Dr. Parsons, the Ward-Coleman Chair in Cancer Research. “And that has pushed the envelope in terms of our knowledge of the biology of cancer and, just as importantly, how we treat it. We’re seeing better outcomes for more and more of our patients and believe cancer rates will continue to come down because of treatments we didn’t have in the past, and more informed prevention.”

Dr. Smith, who is also Chief Medical Officer for the Tisch Cancer Hospital and a Professor of Medicine (Hematology and Medical Oncology), described the dramatic changes in cancer care and treatment, particularly in her specialized field of lung cancer.

“When I finished my fellowship training 12 years ago we had just two drugs for lung cancer, and now there are so many more,” she said. “Patients I treated as a fellow are still alive today thanks to clinical trials for new investigative drugs they were able to enroll in.”

Immunotherapies have carved out many of the greatest gains, while also raising some obstacles for the research community. Immunotherapy refers to treatments that use a person’s own immune system to fight cancer.

“The next frontier is determining which patients are going to have a long-term response to immunotherapy, and how do we overcome the resistance we so often see with these therapies,” said Dr. Parsons. “That’s the biology we still need to figure out, and to that end some of the research we’re most excited about is aimed at helping us better understand the switches in the immune system and how they can be regulated therapeutically.”

Two other areas of research where Dr. Parsons sees great promise are tumor suppressors, which are genes that regulate a cell during cell division, and liquid biopsies, which can detect through a simple blood test at the doctor’s office circulating tumor cells and tumor DNA.

With a strong background in tumor suppressors, he sees great advantage in being able to develop gene therapy or other innovative approaches to restore tumor suppressors, a natural part of the body’s defense mechanism that becomes altered or mutated in almost every type of cancer.

Liquid biopsies, still in early-stage development, could be another significant development. “This idea of being able to catch cancers before they are recognizable is going to ultimately move the needle in improving patient survival,” he said.

Cardinale Smith, MD, PhD

For Dr. Smith, early detection includes more aggressive screening by the health care providers.

“Uptake of lung cancer screening has been slow,” she said. “A lot of the work we’ve been doing at Mount Sinai is connecting with the community to understand what their needs are and how they prefer to partner with us. As a result, we’ve increased mammographies for women to detect breast cancer, and improved colorectal cancer screening for both men and women. Now we need to make the same kind of progress with lung cancer screening.”

She noted that as part of its outreach, Mount Sinai in April 2022 launched the Mount Sinai Robert F. Smith Mobile Prostate Cancer Screening Unit after noticing a high mortality rate for the disease in certain neighborhoods of New York City with a high Black male population. The purchase was funded by a $3.8 million donation from philanthropist Robert F. Smith. This successful effort between the Institute and the Department of Urology has been collecting blood samples to measure PSA levels and referring individuals for follow-up care when a problem is detected.

The nation’s war on cancer formally began with the National Cancer Act of 1971, which established the National Cancer Institute. As for the future of cancer care, Dr. Smith foresees patient care navigation and a palliative care workforce as movements with transformative potential.

Navigators with the ability to compassionately guide people through the often challenging cancer screening and treatment process would be an extremely beneficial allocation of resources, she maintains. So would development of specialized palliative care teams that could provide training and skills to oncologists and other clinicians, including nurses and advanced practice providers.

“We know that palliative care when combined with standard oncologic care can improve patients’ quality of life and mood by decreasing depression,” she said. “It also decreases unnecessary utilization of acute care, such as emergency room visits, hospitalizations, and readmission. Most importantly, it aligns cancer care with the goals and values of the patients, which all of us as clinicians need to hold as sacred in the years ahead.”