AI News, Artificial intelligence model “learns” from patient data to make cancer treatment less toxic

Artificial intelligence model “learns” from patient data to make cancer treatment less toxic

MIT researchers are employing novel machine-learning techniques to improve the quality of life for patients by reducing toxic chemotherapy and radiotherapy dosing for glioblastoma, the most aggressive form of brain cancer.

“We kept the goal, where we have to help patients by reducing tumor sizes but, at the same time, we want to make sure the quality of life — the dosing toxicity — doesn’t lead to overwhelming sickness and harmful side effects,” says Pratik Shah, a principal investigator at the Media Lab who supervised this research.

Rewarding good choices The researchers’ model uses a technique called reinforced learning (RL), a method inspired by behavioral psychology, in which a model learns to favor certain behavior that leads to a desired outcome.

The technique comprises artificially intelligent “agents” that complete “actions” in an unpredictable, complex environment to reach a desired “outcome.” Whenever it completes an action, the agent receives a “reward” or “penalty,” depending on whether the action works toward the outcome.

The approach was used to train the computer program DeepMind that in 2016 made headlines for beating one of the world’s best human players in the game “Go.” It’s also used to train driverless cars in maneuvers, such as merging into traffic or parking, where the vehicle will practice over and over, adjusting its course, until it gets it right.

“Instead, we said, ‘We need to reduce the harmful actions it takes to get to that outcome.’” This represents an “unorthodox RL model, described in the paper for the first time,” Shah says, that weighs potential negative consequences of actions (doses) against an outcome (tumor reduction).

On the other hand, the researchers’ model, at each action, has flexibility to find a dose that doesn’t necessarily solely maximize tumor reduction, but that strikes a perfect balance between maximum tumor reduction and low toxicity.

I can give a quarter dose to this person, half to this person, and maybe we skip a dose for this person.’ That was the most exciting part of this work, where we are able to generate precision medicine-based treatments by conducting one-person trials using unorthodox machine-learning architectures,” Shah says.

“Here, you’re just letting a computer look for patterns in the data, which would take forever for a human to sift through, and use those patterns to find optimal doses.” Schork adds that this work may particularly interest the U.S. Food and Drug Administration, which is now seeking ways to leverage data and artificial intelligence to develop health technologies.

Researchers conducted simulated trials on 50 patients using the machine-learning model designed treatment.

“We kept the goal, where we have to help patients by reducing tumor sizes but, at the same time, we want to make sure the quality of life—the dosing toxicity—doesn’t lead to overwhelming sickness and harmful side effects,” principal investigator Pratik Shah said in the release.

AI Can Make Sure Cancer Patients Get Just Enough (but Not Too Much) Treatment

And those five years can be painful — in an effort to minimize the tumor, doctors often prescribe a combination of radiation therapy and drugs that can cause debilitating side effects for patients.

To create an AI that could determine the best dosing regimen for glioblastoma patients, the MIT researchers turned to a training technique known as reinforcement learning (RL).

First, they created a testing group of 50 simulated glioblastoma patients based on a large dataset of those that had previously undergone treatment for their disease. Then they asked their AI to recommend doses of several drugs typically used to treat glioblastoma [oftemozolomide (TMZ) and a combination of procarbazine, lomustine, and vincristine (PVC)] for each patient at regular intervals (either weeks or months).

According to the researchers, this need to strike a balance between a goal  and the consequences of an action — in this case, tumor reduction and patient quality of life respectively — is unique in the field of RL.

“That was the most exciting part of this work, where we are able to generate precision medicine-based treatments by conducting one-person trials using unorthodox machine-learning architectures.”

New way of predicting kidney function could improve chemotherapy dosing for many cancer patients

Kidneys perform a number of vital functions, including filtering waste and toxins out of the blood, producing vitamin D, and regulating blood pressure.

Determination of the GFR is important because the assessment of kidney function can indicate how a disease is progressing, whether a drug treatment is having adverse side-effects on key bodily functions, and if it is safe to prescribe a drug at a certain dose, a question of particular importance to cancer doctors when prescribing chemotherapy drugs.

“Given how important this measure is in day-to-day clinical practice, we felt that we should provide an evidence-based model for its calculation in this context.” Now, in a study published today in the Journal of Clinical Oncology, the authors describe a new and better way to estimate the GFR, which has been developed using data from a large dataset of over 2,500 patients.

Professor Peter Johnson, Cancer Research UK’s chief clinician, said: “Chemotherapy drugs are very powerful, so having the correct dose makes an enormous difference to how effective they are and how we can avoid unnecessary side effects.  This way of measuring how well a patient’s kidneys are working and how quickly chemotherapy drugs like carboplatin leave the body helps to make our treatments more accurate and better suited to each individual.” ReferenceJanowitz, J et al.

The future of patient-centred cancer care and research

What might the future of personalised cancer patient care look like? Meet Linda, a cancer patient in the not-too-distant future. See how her experience ...

Lower Dose of Chemotherapy Effective for Liver Cancer Patients

Researchers at the University of Pennsylvania report that patients with a common type of liver cancer can start treatment with a lower dose of chemotherapy.

The science of psilocybin and its use to relieve suffering

Leading psychopharmacologist Roland Griffiths discloses the ways that psychedelic drugs can be used to create spiritually meaningful, personally ...

NSCLC Patients May Benefit From Single Dose of SBRT

Anurag Singh, MD, shares new research indicating that non-small cell lung cancer (NSCLC) patients may benefit from a single dose of stereotactic body ...

05 - Cancer Drug Dosage - Interview with Dr. Alex Sparreboom

For additional information visit Every patient and every cancer are different. The goal of Dr

Taking psychedelic drugs to ease death

Psychedelic drugs like MDMA and psilocybin are increasingly being used to treat difficult mental health issues like PTSD, and for palliative care. The field of ...

Brain cancer patients provided new hope with altering drug dosage

"The RESCUE study demonstrated that more continuous treatment with the drug Temodal (TMZ) is a valuable ..

07 - Research In Drug Dose Calculations - Interview with Dr. Alex Sparreboom

For additional information visit Every patient and every cancer are different. The goal of Dr

Calculating Radiation Doses for Tumor Treatment with Learning Algorithms | SciPy 2017 | Roy Keyes

Central to the planning of radiation therapy for cancer treatment is the calculation of radiation dose delivered to a patient. The optimization and subsequent ...

Plant a Seed - Patient Stories and Research Efforts

Drs. Judith Gasson, Lauren Pinter-Brown, Rich Finn and Antoni Ribas discuss research advances at UCLAs Jonsson Comprehensive Cancer Center.