{"id":"CU26281","slug":"automated-machine-learning--CU26281","source":{"id":"CU26281","dataset":"techtransfer","title":"Automated machine learning screening for hospital malnutrition","description_":"<p>This technology is a machine learning model that identifies hospitalized patients at risk of developing malnutrition using multi-parameter clinical data. </p>\r\r<h2>Unmet Need: Streamlined, multi-parameter malnutrition screening tool</h2>\r\r<p>Malnutrition is common in hospitalized patients, leading to longer hospital stays, increased complications, and higher mortality rates. Current screening methods rely on varying criteria and prioritize different parameters, leading to heterogeneity in the identification and diagnosis of malnutrition. There is a need for a standardized, data-driven approach that enable accurate and efficient malnutrition risk assessment.</p>\r\r<h2>The Technology: Automated, real-time malnutrition risk prediction model</h2>\r\r<p>This technology is a machine learning model that computes a malnutrition risk-score using demographic data, vital signs, laboratory results, and clinical information to identify hospitalized patients at high risk of malnutrition. The model risk score (output) is updated daily to provide unbiased, real-time, quantitative monitoring of malnutrition, regardless of age, race, or gender. It can be integrated with existing screening workflows and electronic health record systems to support continuous clinical decision-making. </p>\r\r<p>This technology has been validated in clinical settings.</p>\r\r<h2>Applications:</h2>\r\r<ul>\r<li>Inpatient malnutrition risk screening </li>\r<li>Nutrition care planning and intervention</li>\r<li>Prediction of malnutrition-related complications </li>\r<li>Research tool for studying malnutrition etiology </li>\r</ul>\r\r<h2>Advantages:</h2>\r\r<ul>\r<li>Integrates multi-parameter data for holistic risk assessment </li>\r<li>Provides daily risk updates</li>\r<li>Demonstrates minimal bias across an internal patient population</li>\r<li>Delivers quantitative risk scoring</li>\r<li>Compatible with electronic health record systems </li>\r</ul>\r\r<h2>Lead Inventor:</h2>\r\r<p><a href=\"https://www.linkedin.com/in/neil-kavthekar/\">Neil Kavthekar</a></p>\r\r<h2>Related Publications:</h2>\r\r<h2>Tech Ventures Reference:</h2>\r\r<ul>\r<li><p>IR CU26281</p></li>\r<li><p>Licensing Contact: <a href=\"mailto:techtransfer@columbia.edu\">Joan Martinez</a> </p></li>\r</ul>\r","tags":["Electronic health record","Etiology","Machine learning","Malnutrition","Mortality rate","Risk assessment","Vital signs"],"file_number":"CU26281","collections":[],"meta_description":"ML model integrates multi-parameter data to provide daily, real-time malnutrition risk scores for hospitalized patients.","apriori_judge_output":"{\"scores\":{\"novelty\":4.0,\"potential_impact\":4.0,\"readiness\":4.0,\"scalability\":3.0,\"timeliness\":4.0},\"weighted_score\":4.0,\"risks\":[\"Limited public disclosure distance could affect novelty (not enough independent validation).\",\"Reliance on EHR data may raise interoperability and data quality risks.\",\"Regulatory/compliance considerations for clinical decision support tools not addressed.\"],\"one_sentence_take\":\"Strong near-term potential with solid novelty and readiness, but modest scalability and some regulatory/data-quality risks to address.\"}","inventors":["Hanqing Cao","Joshua Finer","Neil Kavthekar","Sean Yun"],"manager":"Joan Martinez","depts":[],"divs":[],"date_released":"2026-05-29"},"highlight":{},"matched_queries":null,"score":0.0}