Image-guided robotic system for detection and treatment

A system for diagnosing and treating a patient, the system comprising:

• one or more processors;
• a surgical robotic device;
• one or more computer-readable storage media having stored thereon processor-executable instructions that, when executed by said one or more processors, cause the one or more processors to perform a method comprising:
• training, using a first training data set, a machine learning classifier model to detect lesions in magnetic resonance (MR) images of an organ, wherein said machine learning classifier model is implemented as a pipeline of deep learning modules, said deep learning modules including an anatomical segmentation module configured to generate an anatomical map, an abnormal finding detection and segmentation module, and an abnormal finding classification module;
• obtaining one or more MR images of an organ of a patient;
• applying the machine learning classifier model to the obtained one or more MR images of the organ of the patient to obtain an output comprising an anatomical segmentation of an imaging volume and location information and an associated probability of malignancy for a lesion;
• determining, based on the probability of malignancy, whether to perform an MRI-guided biopsy on the lesion of the organ of the patient;
• determining a diagnosis for the identified lesion by applying an AI diagnosis model to the obtained one or more MR images of the organ of the patient;
• determining, based on the diagnosis, a treatment plan for said lesion, said treatment plan including a series of instructions for a user to attach one or more treatment tools to said surgical robot device and control instructions for navigating said surgical robotic device through an in-bore pathway to be positioned to perform an in-bore treatment on the lesion;
• and
• performing, by the surgical robotic device, said in-bore treatment on the lesion based on the treatment plan.