INNOVATIONS

big ideas.

Explore the award-winning innovations inspired by the Patient Safety Technology Challenge. From medication errors to surgical complications, this diverse array of pioneering solutions is aimed at enhancing patient safety and improving healthcare outcomes. Poised at the intersection of technology and patient well-being, each idea represents a potential step towards a safer and more resilient healthcare ecosystem.

Innovations by Problem Category

  • Diagnostic Errors

  • Infections

  • Medication-related

  • Procedure/Surgery-related

  • Patient Care

Diagnostic errors

Failure to establish an accurate and timely explanation of the patient’s condition, leading to misdiagnosis, delayed diagnosis, or missed diagnoses. These errors can have serious consequences for patient safety, resulting in inappropriate treatments, unnecessary procedures, disease progression, and adverse outcomes.

Amanuensis is an AI-enabled physician assistant for automated clinical summarization and question generation. By arming physicians with relevant insights collected from the EHR and patient responses to NLP-generated questions, we empower them to achieve more accurate diagnoses and effective treatment plans. 

TreeHacks

—

Team Deep Sepsis created PhLORENS, an AI-based application for early detection of sepsis in the ICU. They created a demo of the PhLORENS app, which is a physiological learned object response emergency notification system for sepsis detection.

Biomedical Data Science Hackathon 

—

SymptoSecure Plus is a medical profiling tool intended to close the gap between patients and healthcare providers, with a specific emphasis on tackling healthcare disparities experienced by marginalized communities. The digital pre-consultation platform will assist patients in arranging and conveying their medical history, symptoms, and relevant information in a well-structured format.

HackNC

—

DAVE is a passive medical assistant that chimes into patient visits with additional information.

CatHacks

—

Clarity is a multi-modal suite aimed at detecting fatigue in doctors through capturing facial expression in real-time and returning a bounding box around it indicating whether the user is fatigued or active.

HackGT

—

SimplySign uses a machine learning model to analyze hand gestures and then translates American Sign Language (ASL) to English text that is displayed on the screen.

HackHarvard

—

BioFuture believes the future of medicine is personalized, computational, quantitative, and interdisciplinary. The team's idea was an integrated computational facility where clinical staff work in conjunction with multidisciplinary teams focused on computational modeling work to improve care. Examples of their proposed technology include using digital teams in risk assessment and increasing personalized medicine.

Emory University Healthcare Futuring Competition 2040

—

“Telegaming Doc” seeks to address the miscommunication between patient and provider that can lead to poor care through a game-like, online self-evaluation tool.

ASU HEALab Health Innovation

—

Concussion Evaluation Sharing Tool is designed to help emergency medical services (EMS) teams responding to the scene of an accident score concussions. The assessment could be done offline which made it especially useful for rural settings.

Bio-Hack at OSU

—

IPVision won for their idea to transform intimate partner violence (IPV) healthcare by leveraging AI to detect cases and provide personalized treatment in acute care settings.

Cornell Health Hackathon

—

NexusCare provides doctors with one-click access to geriatric research in the EHR system.

Healthcare Hackathon with AI

—

CognitiveKeyboard is an Android keyboard that tracks user typing patterns and uses keystroke data and researcher feedback to predict risks of cognitive decline and mental health conditions.

HackHarvard

—

Elythea developed a machine-learning platform to predict pregnancy complications. Maternal health affects the most women out of any health condition but has yet to face the advances of AI.  

Fowler Global Social Innovation Challenge

—

Inferomics is a platform that connects data scientists and healthcare workers through an accessible website. It enables pathologists to easily get second opinions and improve diagnosis.

HooHacks Ideathon

—

Curis is a project focused on keeping oncologists updated on the optimal course of action for their patients. The team aims to improve the standard process for clinical trial matching which can be inefficient. The AI-powered tool can identify pertinent medical trials tailored to the individual patient's needs based on matching algorithms they created.

HackUTD

—

MediBuddy leverages machine learning models, to analyze symptoms, body measurements, and more inputs by doctors to generate accurate, personalized suggestions on diagnoses and treatments based on previous case data.

LAHacks

—

MedTap utilizes NFC technology to link to a user’s medical profile, which displays important information including medical history, allergies, and emergency contacts. This innovation will reduce the amount of time it takes medical professionals to obtain vital patient information in emergencies.

HopHacks

—

INFECTIONS

Healthcare-associated infections (HAIs) are infections that patients acquire while receiving medical care in a healthcare facility. These infections can result from a variety of factors, including exposure to pathogens in the healthcare environment, invasive procedures, use of medical devices, and interactions with healthcare personnel. Examples of common infections include respiratory infections, surgical site infections, sepsis and C. diff.

Elivio Health is focused on the $8.2B Inpatient Surgical Procedures market to reduce hospital readmission rates due to surgical site infections. With Elivio Health technology monitoring the patient’s incision wound 24/7, doctors and nurses have a direct line to critical health metrics, while Elivio also keeps patients informed of their status from the comfort of their home.

Columbia-CareOne Healthcare Innovation Challenge

—

Team AERO developed a device to detect the risk of pressure injury in patients. The idea uses a standardized amount of compressed air applied to a patient's skin in a controlled and consistent manner, then uses a high-speech camera to capture how skin and tissue respond. The images from this response are processed to find mechanical properties of skin and underlying tissue to identify early signs of skin/tissue deterioration. 

Healthcare Innovation Pitch Program

—

Uroglo focuses on the early detection of catheter-associated urinary tract infections (CAUTI). Their novel solution coats the inside of urine collection bags with a hydrogel infused with bacteriophages that will glow under UV when harmful bacteria that cause CAUTI are identified. 

Healthcare Innovation Sprint

—

VAPGuard is a novel mouthguard that clips around an endotracheal tube for ventilated patients. Using a specific wavelength of blue light, the light technology, inbuilt within the mouthguard, can kill the bacteria that colonizes inside a patient's mouth. 

University of Rochester Patient Safety Technology Challenge 

—

Femnostics addressed female reproductive health issues, by creating a device to detect vaginal infections more efficiently.

Cornell Health Hackathon

—

Anti-Sepsis created a health tracker device to judge a patient’s sepsis susceptibility using standard sepsis testing metrics including complete blood count, lactate level, and other biochemical markers. Based on the metrics, the device will provide patients with therapy and recovery plans on the screen.

Hacking Health

—

H-AI-HAI is developing machine learning models to automate healthcare-associated infection (HAI) surveillance, aiming to replace the current manual method with a more efficient, cost-effective system. This innovation could allow infection prevention teams to focus more on implementing preventive measures rather than on the labor-intensive process of measuring HAIs.

Big Ideas Competition at Washington University of School of Medicine St. Louis 

—

Medication-Related

Medication errors are one of the most common types of preventable adverse events in healthcare settings, posing significant risks to patient safety. A medication error can occur at any stage of the medication use process, from prescribing and transcribing to dispensing, administering, and monitoring. The top five types of harm events include delirium or other change in mental status, hypotension, acute kidney injury, excessive bleeding, and hypoglycemia.

Pill Identifier allows a user to take a picture or upload a file of a pill and in return get which pill is most likely classified as from our data set. Machine learning algorithm and a neural network model were used to compare images against our dataset. The model would look for similarities in shape, size, and color, and return the name of the pill that most accurately represents input as well as the prediction percentage that tells the user how accurate the model is. 

The Pitt Challenge

—

Boosting Patient Safety noticed that adverse drug reactions have been doubling in the last 10 years and wanted to see what kind of insights they could identify that could help patients and or clinicians avoid ADRs. Using the FDAs ADR database they were able to find that two medications had a higher incidence of harm - Enbrel and Humira.

Columbia Data Science Society Hackathon

—

ManagMed is a consumer-facing mobile app designed for patients to use after they visit their doctor. The app would help patients: look up their insurance coverage and provide suggestions for generic alternatives, set up reminders to take medications based on a scan of their medication labels, alert them about drug-drug interactions, and embed a chatbot to answer health questions.

MakeUC

—

Crimson Medical Solutions’ IV Manager is an improved IV labelling system which can improve nurse workflows and reduce medication error.

NursesKnow Patient Safety Technology Challenge

—

Pillendar is a configurable pillbox that can provide cues to users based on their needs along with an accompanying app that acts as a mediator between patients and their healthcare team.

Purdue’s Design and Innovation Competition

—

MediTrack is an app that uses unique patient data along with dosage guidelines to enable healthcare providers to verify the safety of medications for patients prior to prescribing them. 

DivHacks

—

Prescription Evaluation uses GPT4 and openFDA. The demo of a prescription checker used AI to reduce human oversight and improve effectiveness. 

HackAuton

—

Rx Tracking created an app prototype designed to empower patients to easily document and keep track of their prescriptions so that they could be a reliable source of truth when inevitably a clinician asks, “What medications are you taking?”

MedHacks

—

RPillPal is a biometric pill dispenser to avoid opioid overdoses after patients are discharged from the hospital and ensure only that person living in the home has access to that medication. RPillPall is intended for home use that uses fingerprints for patient identification. It also allows for direct connectivity with healthcare providers to facilitate progress monitoring and timely treatment adjustments.

HackRPI

—

Pharmany is a personalized patient drug portfolio that generates drug-drug interaction reports for medical professionals.

Hacking Health

—

MediCheck created a system to augment and enhance drug selection safety by providing an additional check.

CatHacks

—

Medicine Dispensary aims to facilitate the pill taking process by providing essential details about medication, including its type and the scheduled times for dosage, automatically dispensing the correct medication and dosage at the correct time.

ShellHacks

—

Althea is an app that enables users to track their medications with a daily checklist, along with symptom tracking and cross-referencing medication side effects with symptoms a user experienced and entered into the app.

HackHarvard

—

MemorAIs is a platform where users can scan the prescription bottle label’s intake directions to generate and download a .ics (iCalendar) file pre-configured with intake frequency, duration, and time information straight to their device. 

DivHacks

—

PHARMABOT is a pharmaceutical robot that checks and packages pills to prevent medication errors in the UK.

TeenTech Award: Global Challenge on Patient Safety

—

“Decoding Devices: Improving Patient Safety with MAUDE”  solution focused on getting natural language text to collect all the problems related to a specific medical device into one place so action to avoid future harm can be taken. The solution extracts important text about device malfunctions using LexRank extractive summarization, then summarizing the results using GPT abstractive summarization.

Columbia Data Science Society Hackathon

—

SafeRx created a mechanism for physicians to assess adverse drug reactions, before they occur using natural language processing, computer vision, and AI models.

Hack-a-Startup

—

RocketHealth is a co-pilot for home health nurses. It works to counter adverse medical effects and provide medication reconciliation and automation.

Cornell Health Hackathon

—

Snapill is a computer vision app that tracks medication related information by allowing a user to scan their medication vial and automatically generating important metadata like prescription name, expiration date, dosage, and possible times to take the pill, along with reminders. The chat feature uses Cerebras AI to enable users to interact with their medications and get more information, like incompatibility between different medications.

PennApps XXV

—

Procedure/ Surgery-Related

Harms that are commonly experienced during procedures or surgeries. Common examples of related harms include hypertension during surgery, excessive bleeding, and embolisms (i.e., vascular and fat embolisms).

MedVisio-on acts as a supplementary cognitive resource and additional set of eyes for perioperative nurses which allow them to better monitor equipment, alleviate their mental workload, and enable them to concentrate on the surgical procedure. Simultaneously, it was designed to diminish the avoidable risk of Retained Surgical Bodies (RSBs), like scalpels and sponges.

HackSC

—

Countability employs computer vision and object tracking to take real-time inventory of surgical implements during procedures. It sends alerts when objects are unaccounted for to reduce medical harm from surgical tools being left inside of patients.

HopHacks

—

SurgiScan leverages computer vision technology to detect and track surgical tools during procedures, reducing surgical errors, improving patient safety, and preventing malpractice.

HopHacks

—

SafeOR uses modern technologies such as computer vision, NLP, and machine learning to bring visibility to patient safety and the need for tech-enabled solutions to reduce medical errors. 

TAMUhack

—

MedForward redesigned surgical pedals with haptic feedback to reduce the risk of surgeons pressing the wrong pedal, saving time and resources while reducing potential complications.

University of Rochester Patient Safety Technology Challenge

—

Cedar created a Portable Ultrasound Imager for Carotid Artery Stenosis.

Cornell Health Hackathon

—

OnPoint Ventilation’s device, the Bronchosleeve will enhance lung ventilation by improving ease of insertion and reducing life-threatening complication.

Fowler Global Social Innovation Challenge

—

M2 designed an AI-directed training simulator for placement of drains into hydrocephalus patients and traumatic brain injuries.

University of Rochester Patient Safety Technology Challenge

—

SurgiSafe is a wearable device designed to monitor wounds by detecting moisture, temperature, and pH levels in real time. This device provides crucial data, enabling early alerts for patients and healthcare providers, which can help prevent serious infections.

PennApps XXV

—

patient care

Patient care pertains to the daily care of patients, which is often performed by nurses. Examples of common harms associated with patient care include pressure injuries, falls, and venous thromboembolism (blood clots).

Wireless Fall Alert System is a working prototype of a belt that a patient could wear to detect falls and automatically send a text message to a designated third party. This concept included sensors in the belt to collect data that would be used to develop a capability to detect when a patient was at risk of falling thereby moving into the fall prevention realm. 

MedHacks

—

IMPACTT- Stem Cell Transplant Module guides a patient from the intake session through customizable educational and survey sessions to transplant patients 

Penn Nursing Innovation Accelerator 

—

The Pegasus Safety Breakaway System elevates the universal utility of common medical tubing by reducing risks to patient safety, hospital liability, and complications inherent in current standard of care.

Bench to Bedside Competition

—

Team SpeakNOW developed a virtual reality platform to reduce patient stress in clinical settings.

Healthcare Innovation Pitch Program

—

Medcrypt products are optimized for medical device manufacturers to build safe, secure, innovative medical devices faster while meeting U.S. Food & Drug Administration cybersecurity requirements. Medcrypt assists healthcare technology firms in guaranteeing the security of their medical devices by offering cybersecurity solutions and advisory services to streamline the introduction of lifesaving connected technologies into the market.

Digital Health Hub Awards

—

NurseWav created a tech-enabled solution to address alarm fatigue for nurses in non-acute post-surgery units in hospitals.

ASU Social Innovation Start-up Lab

—

LIFE-line is a device that detects falls and notifies either a registered family member or the local emergency services depending on the severity of the fall. It has an OLED display for text, intended for a speech-to-text implementation to help either overcome language barriers or communication disabilities.

IDEA Hacks

—

Current Care created a device to prevent pressure ulcers through electrical muscle stimulation.

Cornell Health Hackathon

—

Scout, is an innovative device to improve workflow efficiency during high-stress situations using an intuitive hands-free vein visualizer.

Purdue’s Design and Innovation Competition

—

Vigil is a mobile app that allows patients in hospitals or nursing homes to send requests to nurses, replacing the nurse call light. It has a patient portal, as well as a nurse portal, acting as a one-stop-shop for nurses.

DubHacks

—

Curis is a project focused on keeping oncologists updated on the optimal course of action for their patients. The team aims to improve the standard process for clinical trial matching, which can be inefficient. The AI-powered tool can identify pertinent medical trials tailored to the individual patient's needs based on matching algorithms they created. 

HackUTD

—

SMARTRACH is developing a device that wirelessly monitors airflow through a tracheostomy tube and syncs with an alert system that notifies caregivers if obstructions occur. 

Hollomon Health Innovation Challenge 

—

Team “VTE” focused on patient non-compliance with use of Intermittent Pneumatic Compression devices by envisioning an IPC that was more comfortable to wear and less expensive to manufacturer. Their solution would increase patient safety by increasing their compliance with the therapy.

MedHacks

—

DEMENSURE is an innovative implant to monitor and care for dementia patients in the community.

TeenTech Award: Global Challenge on Patient Safety

—

AEyes is a cross-platform mobile app for the visually impaired integrating computer vision as the basis of their tech. The app acts as the eyes of a person and takes an image of their surroundings to better understand the environment around them and tells the user, using AI-based text-to-speech, what is around them and what is potentially hazardous.

VandyHacks

—

HealMate Pro aims to help monitor patients at the bedside by tracking patient movements to visualize and report potential issues in real time. The team chose to focus on patients that suffer from sporadic behavior while unconscious, specifically tailored to monitor patients with epilepsy. They created a motion tracker for patient movements. The movements are visualized in real time which can be accessed in remote locations.

Hack the Nest

—

Meddit creates exclusive and anonymous spaces for hospitals’ medical communities to share their experiences around medical harm. Through Meddit, staff can be inspired to report errors rather than hide them and spread awareness of some of the most common medical errors to prevent other medical workers from committing the same mistakes in the future.

IrvineHacks

—

Code Clock won for their device to optimize care and communication among code team members in the setting of an in-hospital cardiac arrest. In-hospital cardiac arrest can still be as high as 70-90%. Code Clock’s device aims to improve patient survival by improving communication.

Bench to Bedside

—

TL;Dr. aims to reduce medical error that occurs at handoff and discharge stages of patient care. This is done by standardizing and automating the reporting procedure to provide higher quality communication from doctor-to-doctor and doctor-to-patient. OpenAI's GPT-3 was used to summarize lengthy handoff notes into a standard format called IPASS.

HackMIT

—

NephSol is a brain blood flow monitoring device for dialysis patients that will provide real-time alerts and continuous monitoring designed to curb common risks that patients encounter during dialysis. Their product is a non-invasive, functional, and uses a near infrared spectroscopy headset that can provide continuous monitoring, real-time alerts, and comfort for patients.   

Edward J. Aldag Jr. Student Business Plan Competition 

—

Octila Health addressed how mental health patients are not adequately equipped and educated to thrive after discharge. Their app enables home patient monitoring and education, while keeping their care team in the loop.

MIT Hacking Medicine’s Grand Hack

—

Patient Watch  is an automatic monitoring and warning system solution to alleviate burden on frontline workers. It monitors patients’ conditions through web cameras and alerts staff to suboptimal conditions.

HackDuke

—

Man Overboard focused on a safety system crafted to identify passengers going “overboard,” promptly alerting ship crews in real-time. Employing computer vision and instantaneous communication, their system consistently surveils the ship’s surroundings, promptly notifying the crew upon detecting a potential overboard situation.

HackNJIT

—

Proxie Health is a telecare platform that aims to give providers data they need and help families manage home care using human expertise and AI.

CTA Foundation Pitch Competition

—

Reel Free created a medical device that retracts 50-feet of oxygen tubing to prevent falls among supplemental oxygen users.

Arizona State University’s Innovation Open (ASUio)

—

Nebu is an innovative nebulizer for asthmatic patients that helps reduce the number of patients harmed from complications

Purdue’s Design and Innovation Competition

—

Kalogon’s product, the Orbiter Smart Cushion, redistributes pressure points and improves blood flow to prevent the risk of pressure injuries.  The smart cushion works to reduce preventable harm by first using pressure sensors to detect applied pressure on the seat, and then a control system takes the information and adjusts the seat’s pressure as necessary. The cushion comes with a Bluetooth-enabled app that users and their care team can utilize to personalize the product. 

South by Southwest (SXSW)