diff --git a/docs/source/getting_started/tutorials/mednist_app.md b/docs/source/getting_started/tutorials/mednist_app.md
index 539d98fe..fa7e3605 100644
--- a/docs/source/getting_started/tutorials/mednist_app.md
+++ b/docs/source/getting_started/tutorials/mednist_app.md
@@ -88,6 +88,14 @@ monai-deploy package examples/apps/mednist_classifier_monaideploy/mednist_classi
     --model classifier.zip \
     -l DEBUG
 
+# For AMD GPUs, nvidia-docker is not required. Use --base [base image] option to override the docker base image.
+# Please see https://hub.docker.com/r/rocm/pytorch for rocm/pytorch docker images.
+monai-deploy package -b rocm/pytorch:rocm5.4.1_ubuntu20.04_py3.7_pytorch_1.12.1 \
+    examples/apps/mednist_classifier_monaideploy/mednist_classifier_monaideploy.py \
+    --tag mednist_app:latest \
+    --model classifier.zip \
+    -l DEBUG
+
 # Run the app with docker image and input file locally
 monai-deploy run mednist_app:latest input output
 cat output/output.json
diff --git a/docs/source/getting_started/tutorials/monai_bundle_app.md b/docs/source/getting_started/tutorials/monai_bundle_app.md
index bc47b1b5..986e1ff2 100644
--- a/docs/source/getting_started/tutorials/monai_bundle_app.md
+++ b/docs/source/getting_started/tutorials/monai_bundle_app.md
@@ -66,6 +66,10 @@ monai-deploy exec ../examples/apps/ai_spleen_seg_app/app.py -i dcm/ -o output -m
 # Please see https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#docker to install nvidia-docker2.
 monai-deploy package examples/apps/ai_spleen_seg_app --tag seg_app:latest --model model.ts -l DEBUG
 
+# For AMD GPUs, nvidia-docker is not required. Use --base [base image] option to override the docker base image.
+# Please see https://hub.docker.com/r/rocm/pytorch for rocm/pytorch docker images.
+monai-deploy package -b rocm/pytorch:rocm5.4.1_ubuntu20.04_py3.7_pytorch_1.12.1 examples/apps/ai_spleen_seg_app --tag seg_app:latest --model model.ts -l DEBUG
+
 # Run the app with docker image and input file locally
 monai-deploy run seg_app:latest dcm/ output
 ```
diff --git a/docs/source/getting_started/tutorials/multi_model_app.md b/docs/source/getting_started/tutorials/multi_model_app.md
index 44b2bcbd..307cb6f1 100644
--- a/docs/source/getting_started/tutorials/multi_model_app.md
+++ b/docs/source/getting_started/tutorials/multi_model_app.md
@@ -66,6 +66,10 @@ monai-deploy exec ../examples/apps/examples/apps/ai_multi_ai_app/app.py -i dcm/
 # Please see https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#docker to install nvidia-docker2.
 monai-deploy package -b nvcr.io/nvidia/pytorch:22.08-py3 examples/apps/ai_multi_ai_app --tag multi_model_app:latest --model multi_models -l DEBUG
 
+# For AMD GPUs, nvidia-docker is not required. Use --base [base image] option to override the docker base image.
+# Please see https://hub.docker.com/r/rocm/pytorch for rocm/pytorch docker images.
+monai-deploy package -b rocm/pytorch:rocm5.4.1_ubuntu20.04_py3.7_pytorch_1.12.1 examples/apps/ai_multi_ai_app --tag multi_model_app:latest --model multi_models -l DEBUG
+
 # Run the app with docker image and input file locally
 monai-deploy run multi_model_app:latest dcm/ output
 ```
diff --git a/docs/source/getting_started/tutorials/segmentation_app.md b/docs/source/getting_started/tutorials/segmentation_app.md
index 2905729d..9ee72515 100644
--- a/docs/source/getting_started/tutorials/segmentation_app.md
+++ b/docs/source/getting_started/tutorials/segmentation_app.md
@@ -76,6 +76,10 @@ python examples/apps/ai_spleen_seg_app/app.py -i dcm/ -o output -m model.ts
 # Please see https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#docker to install nvidia-docker2.
 monai-deploy package examples/apps/ai_spleen_seg_app --tag seg_app:latest --model model.ts -l DEBUG
 
+# For AMD GPUs, nvidia-docker is not required. Use --base [base image] option to override the docker base image.
+# Please see https://hub.docker.com/r/rocm/pytorch for rocm/pytorch docker images.
+monai-deploy package -b rocm/pytorch:rocm5.4.1_ubuntu20.04_py3.7_pytorch_1.12.1 examples/apps/ai_spleen_seg_app --tag seg_app:latest --model model.ts -l DEBUG
+
 # Run the app with docker image and input file locally
 monai-deploy run seg_app:latest dcm/ output
 ```
diff --git a/monai/deploy/packager/util.py b/monai/deploy/packager/util.py
index 3db72128..2249775b 100644
--- a/monai/deploy/packager/util.py
+++ b/monai/deploy/packager/util.py
@@ -42,7 +42,10 @@ def verify_base_image(base_image: str) -> str:
         str: returns string identifier of the dockerfile template to build MAP
         if valid base image provided, returns empty string otherwise
     """
-    valid_prefixes = {"nvcr.io/nvidia/cuda": "ubuntu", "nvcr.io/nvidia/pytorch": "pytorch"}
+    if "rocm" in base_image:
+        valid_prefixes = {"rocm/pytorch": "ubuntu"}
+    else:
+        valid_prefixes = {"nvcr.io/nvidia/cuda": "ubuntu", "nvcr.io/nvidia/pytorch": "pytorch"}
 
     for prefix, template in valid_prefixes.items():
         if prefix in base_image:
@@ -91,10 +94,12 @@ def initialize_args(args: Namespace) -> Dict:
         if not dockerfile_type:
             logger.error(
                 "Provided base image '{}' is not supported \n \
-                          Please provide a Cuda or Pytorch image from https://ngc.nvidia.com/ (nvcr.io/nvidia)".format(
+                        Please provide a ROCm or Cuda based Pytorch image from \n \
+                        https://hub.docker.com/r/rocm/pytorch or https://ngc.nvidia.com/ (nvcr.io/nvidia)".format(
                     args.base
                 )
             )
+
             sys.exit(1)
 
     processed_args["dockerfile_type"] = dockerfile_type if args.base else DefaultValues.DOCKERFILE_TYPE
diff --git a/monai/deploy/runner/runner.py b/monai/deploy/runner/runner.py
index ee3e183b..d521bf30 100644
--- a/monai/deploy/runner/runner.py
+++ b/monai/deploy/runner/runner.py
@@ -20,6 +20,7 @@
 from typing import Tuple
 
 from monai.deploy.runner.utils import get_requested_gpus, run_cmd, verify_image
+from monai.deploy.utils.deviceutil import has_rocm
 
 logger = logging.getLogger("app_runner")
 
@@ -87,7 +88,8 @@ def run_app(map_name: str, input_path: Path, output_path: Path, app_info: dict,
     # Use nvidia-docker if GPU resources are requested
     requested_gpus = get_requested_gpus(pkg_info)
     if requested_gpus > 0:
-        cmd = "nvidia-docker run --rm -a STDERR"
+        if not has_rocm():
+            cmd = "nvidia-docker run --rm -a STDERR"
 
     if not quiet:
         cmd += " -a STDOUT"
@@ -160,12 +162,13 @@ def pkg_specific_dependency_verification(pkg_info: dict) -> bool:
     """
     requested_gpus = get_requested_gpus(pkg_info)
     if requested_gpus > 0:
-        # check for nvidia-docker
-        prog = "nvidia-docker"
-        logger.info('--> Verifying if "%s" is installed...\n', prog)
-        if not shutil.which(prog):
-            logger.error('ERROR: "%s" not installed, please install nvidia-docker.', prog)
-            return False
+        if not has_rocm():
+            # check for nvidia-docker
+            prog = "nvidia-docker"
+            logger.info('--> Verifying if "%s" is installed...\n', prog)
+            if not shutil.which(prog):
+                logger.error('ERROR: "%s" not installed, please install nvidia-docker.', prog)
+                return False
 
     return True
 
diff --git a/monai/deploy/utils/deviceutil.py b/monai/deploy/utils/deviceutil.py
new file mode 100644
index 00000000..6a62e33b
--- /dev/null
+++ b/monai/deploy/utils/deviceutil.py
@@ -0,0 +1,36 @@
+# Copyright 2023 MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import subprocess
+
+
+def has_rocm():
+    """Return True if ROCm is installed and GPU device is detected.
+
+    Args:
+
+    Returns:
+        True if ROCm is installed and GPU device is detected, otherwise False.
+    """
+    cmd = "rocminfo"
+    try:
+        process = subprocess.run([cmd], stdout=subprocess.PIPE)
+        for line_in in process.stdout.decode().splitlines():
+            if "Device Type" in line_in and "GPU" in line_in:
+                return True
+    except Exception:
+        pass
+
+    return False
+
+
+if __name__ == "__main__":
+    print(has_rocm())