A mother whose son had suffered from chronic pain for three years and seen 17 doctors without a diagnosis turned to artificial intelligence for help. ChatGPT, an AI chatbot, suggested a potential diagnosis that was confirmed by a neurosurgeon.
Alex, the 4-year-old boy, had been experiencing chronic pain in his teeth and back. His mother, Courtney, took him to see a variety of specialists, including dentists, orthodontists, and neurologists. But none of them could figure out what was wrong.
Alex’s symptoms were confusing. He was grinding his teeth, and doctors thought it was related to dental issues or airway obstructions. But no single specialist could connect the dots.
One doctor suggested Chiari malformation, a condition affecting the brain and spine. But again, it was a dead end. Courtney was desperate for answers.
Finally, she turned to ChatGPT. She entered all the medical information she had on Alex, from MRI notes to symptoms. ChatGPT suggested tethered cord syndrome, a rare neurological condition in which the spinal cord is abnormally attached to the spinal column.
Courtney was skeptical at first, but she decided to research the condition further. She found a Facebook group for parents of children with tethered cord syndrome, and she was surprised by how many of Alex’s symptoms matched the symptoms of other children in the group.
So, she thought, “Why not give it a shot?” She took Alex to a neurosurgeon who’s an expert in tethered cord syndrome. The doc looked at Alex’s MRI scans and said, “Yep, ChatGPT is onto something! (agreeing with ChatGPT’s diagnosis)”
Alex underwent surgery to correct his tethered cord syndrome, and he is now recovering well.
Courtney shared her story on social media, hoping to raise awareness about tethered cord syndrome and the potential of AI in healthcare. Her posts resonated with many parents who were also struggling to find a diagnosis for their children’s unexplained symptoms.
AI’s Impact on Healthcare is Growing
This is not an isolated case. Across the globe, AI is being used to assist in diagnosing complex medical conditions, personalizing treatment plans, predicting health issues before they become serious, and even performing surgeries.
For example, Oracle has integrated generative AI with voice command in its new Clinical Digital Assistant built on the proven Oracle Digital Assistant platform.
It enables physicians to give their full attention to patients while dramatically simplifying administrative tasks. The multimodal voice and screen-based assistant participates in the appointment using generative AI to automate note-taking and to propose context-aware next actions, such as ordering medication or scheduling labs and follow-up appointments.
It also responds to conversational voice commands from providers who can ask questions such as ‘show me the patient’s latest MRI results,’ to look up elements of a patient’s EHR record during an appointment.
The information and images are then delivered in a relevant order that helps the physician gain insight into the appropriate treatment path without requiring a multi-menu, multi-step interaction with the EHR.
On the global stage, an Indian AI healthcare model, August AI, has demonstrated exceptional capabilities, surpassing the performance of prominent AI counterparts such as GPT-4 and MedPaLM.
Developed by the Bangalore-based health AI startup Beyond, August AI aims to democratize access to high-quality health information.
August AI scored 94.8% on the US Medical Licensing Examination (USMLE), outperforming MedPaLM which scored 85%, and GPT-4 which scored 87.8%.
The model is multimodal, accepting input in various formats such as audio, text, and PDF Lab, and delivers the output in text format. It currently aids people in handling physical and mental health concerns and simplifying complex health procedures.
The platform is currently free for use on Whatsapp. The company believes that Whatsapp is a great way to access the platform, and their key focus is to enable natural conversation on Whatsapp.
While Google is building for the US, August’s focus on India and its empathetic conversation will be key differentiators. The model has been developed by a team of engineers, data scientists, and doctors, and aims to bridge the gap between doctors and patients.
In addition to this, a study conducted at the Postgraduate Institute of Medical Education and Research (PGIMER) in Chandigarh, India, and the Indian Institute of Technology (IIT), New Delhi, has shown that an AI-based approach can detect gallbladder cancer as accurately as experienced radiologists.
The team developed and validated a deep learning (DL) model for gallbladder cancer (GBC) detection using abdominal ultrasound and compared its performance with radiologists.
The DL model was trained on a dataset of 233 patients, validated on 59 patients, and tested on 273 patients. In the test set, the DL model had a sensitivity of 92.3%, a specificity of 74.4%, and an area under the receiver operating characteristic curve (AUC) of 0.887 for detecting GBC, which was comparable to both radiologists.
The DL-based approach showed high sensitivity and AUC for detecting GBC in the presence of stones, contracted gallbladders, small lesion size (less than 10 mm), and neck lesions, which were also comparable to the radiologists.
In the future, AI-powered tools will help doctors to diagnose diseases earlier, develop more effective treatments, and predict and prevent health problems before they occur. AI will also be used to make healthcare more affordable and accessible for everyone.
References from:
https://analyticsindiamag.com/this-indian-ai-healthcare-model-outperformed-gpt-4-and-medpalm/
