When discussing examples of AI in healthcare, our minds may jump to automated call centers and schedule reminders, chatbots and even robotic surgeries. One crucial example of AI in medicine, however, is its application to clinical workflows and how it can help improve patient outcomes. In this blog post, we’re going to dive into 10 possible use case examples.
Use Case: A 49-year-old male is referred for a head CTA following a long period of headaches. In this instance, AI can analyze the CTA image, which reveals a saccular aneurysm located at the bifurcation of the middle cerebral artery. The aneurysm measured 5 mm in diameter, with no signs of rupture. AI can raise an alert to notify the physician of a suspected aneurysm. The physician can then confirm the brain aneurysm and determine the best treatment and follow-up for that patient. In this instance, AI has ensured this finding was triaged for the care team in real-time, offering specialists in the health system an opportunity to deliver optimal care.
Use Case: A 42-year-old male fell off a ladder, resulting in severe abdominal pain. The ED team found a limited range of motion of the cervical spine, thus adding a CT of the cervical spine to the abdominal examination. The patient arrived at the scanner without a collar. In this instance, AI flags the subtle cervical spine fracture, which the radiologist can confirm. This is just one benefit of AI in healthcare, as it opens up the care team to initiate a spinal precaution protocol intervention.
Use Case: A 75-year-old female on anticoagulation arrives at the ED after falling in the bathtub a day earlier. AI flags a small right parietal subdural hematoma and prioritizes the case in the mobile care coordination app and radiology worklist. This example of AI in medicine gives clinicians a leg up on expediting care.
Use Case: A 60-year-old male with hypertension and smoking history presents to the ED with sudden right-sided weakness and speech difficulty. The ED noted right hemiparesis and facial droop with an NIHSS of 12, suggesting a moderate to severe stroke. A NCCT and CTA were ordered and performed. A huge benefit for AI in stroke care is its ability to first analyze that the NCCT was negative for a hemorrhage. On the CTA, the algorithm flags a suspected left middle cerebral artery occlusion at the M2 segment, alerting the stroke care team and neurointerventionalist. The care team may order intravenous thrombolytics while the neuro interventional physician reviews patient images and EHR data, determining if the patient should proceed for mechanical thrombectomy.
Use Case: A 49-year-old female, otherwise healthy, arrived after a long flight with atypical chest pain and shortness of breath. In this case, AI can flag a subtle subsegmental PE that drastically reduced waiting and the reading times for the study, also opening opportunities to coordinate care with the Pulmonary Embolism Response Team.
Use Case: A 36-year-old male, scanned during a routing restaging following chemotherapy. An incidental pulmonary embolism algorithm can help flag a subtle, but clinically significant, pulmonary embolus in the right middle lobe, leading to reduced turnaround time in radiologist diagnosis and in the notification of the downstream care team.
Use Case: A 53-year-old male with hypertension presents to the ER with moderate chest pain. The patient has normal EKG and is waiting for troponin lab results. In the meantime, the patient goes to CT. A benefit of AI in treatment aortic conditions is that it can flag a suspected dissection and prioritize the scan to the radiologist. The radiologist then confirms the dissection in their PACS and forwards the findings to the intervention team through a mobile app. By the time the patient is returned to the ED, clinicians and radiologists are aware of the dissection and call it out to the surgical staff.
Use Case: A 65-year-old male, ex-smoker with hypertension and hyperlipidemia, experienced intermittent abdominal pain without gastrointestinal symptoms. An examination found a stable blood pressure and a pulsatile abdominal mass, suggestive of an abdominal aortic aneurysm (AAA). A contrast CT was run, and AI can mark the study, identifying that the measurements (5.2 x 4.4 cm) are above the site configured 3 cm threshold. The radiologist then adds the details of a suspected AAA that appears not to have ruptured or have a dissection, but did not mention it to the vascular care team. The AI can then pick up the mention of the AAA in the report and notify the vascular care team to further evaluate and decide if an intervention is required.
Use Case: A 45-year-old male undergoes a non-contrast CT after a car accident. Though it’s not what the radiologists and ED physicians are looking for, AI could potentially flag the possibility of a high level of coronary artery calcification, an important indicator of cardiovascular health. The radiologist reviews the CAC series in PACS, adds the CAC details to the radiologist report and the patient is referred to a cardiologist for further management.
Use Case: A 29-year-old female, post-motor vehicle accident, is admitted with multiple injuries. She is hemodynamically stable with pain localized to the left upper quadrant, working diagnosis was a splenic trauma. Fortunately, AI flagged suspected rib fractures, helping reduce the radiologist turnaround time and prioritizing the patient for additional evaluation with the orthopedic care team.
The impact of AI in the healthcare domain, in many ways, is only seeing its initial impact. With an ever growing pool of evidence suggesting the clinical efficacy of AI, use cases like the ones outlined above are bound to multiply. AI powered healthcare is an undeniable future.
The above use cases are just some examples of AI in healthcare, with plenty more clinical use cases and benefits yet to be seen. Learn more about enterprise-wide AI and how it effectively helps health systems overcome some of the challenges of AI adoption.
*The information presented in this blog are not specific to Aidoc technology and are intended only for educational purposes about clinical AI broadly. For information on Aidoc’s cleared products indications for use; safe and correct usage and risk information, please refer to Aidoc’s 510(k) premarket notifications on the FDA’s website and to the product’s User Guide.
Aidoc experts, customers and industry leaders share the latest in AI benefits and adoption.
Explore how clinical AI can transform your health system with insights rooted in real-world experiences.
Learn how to go beyond the algorithm to develop a scalable AI strategy and implementation plan.
Explore how Aidoc can help increase hospital efficiency, improve outcomes and demonstrate ROI.
