diff --git a/.github/workflows/build-test.yaml b/.github/workflows/build-test.yaml
index f4b217a..8708cdb 100644
--- a/.github/workflows/build-test.yaml
+++ b/.github/workflows/build-test.yaml
@@ -221,6 +221,8 @@ jobs:
           max_attempts: 5
           command: |
             python -c "import geant4_python_application as g4; g4.install_datasets()"
+            # Check that the datasets are installed correctly
+            python -c "import geant4_python_application as g4; g4.files.datasets.check_datasets(throw=True)"
 
       - name: Run tests
         run: |
diff --git a/src/geant4_python_application/files/datasets.py b/src/geant4_python_application/files/datasets.py
index a7a6095..05ee991 100644
--- a/src/geant4_python_application/files/datasets.py
+++ b/src/geant4_python_application/files/datasets.py
@@ -17,11 +17,9 @@
 url = "https://cern.ch/geant4-data/datasets"
 
 
-# It is discouraged to use the package directory to store data
+# It is discouraged to use the python package directory to store data
 # data_dir = os.path.join(os.path.dirname(__file__), "geant4/data")
 # another idea is to use 'platformdirs' to store data in a platform-specific location
-
-
 def data_directory() -> str:
     return os.path.join(
         geant4_python_application.application_directory(),
@@ -32,7 +30,11 @@ def data_directory() -> str:
     )
 
 
-# the datasets versions should be updated with each Geant4 version
+def geant4_data_directory() -> str:
+    return os.environ["GEANT4_DATA_DIR"]
+
+
+# the datasets versions should be updated with each Geant4 version (remember to update the checksum too!)
 # https://geant4.web.cern.ch/download/11.2.2.html#datasets
 Dataset = namedtuple("Dataset", ["name", "version", "filename", "env", "md5sum"])
 
@@ -42,7 +44,7 @@ def data_directory() -> str:
         version="4.7.1",
         filename="G4NDL",
         env="G4NEUTRONHPDATA",
-        md5sum="b001a2091bf9392e6833830347672ea2",
+        md5sum="54f0ed3995856f02433d42ec96d70bc6",
     ),
     Dataset(
         name="G4EMLOW",
@@ -155,6 +157,9 @@ def _download_extract_dataset(dataset: Dataset, pbar: tqdm):
         for chunk in iter(lambda: f.read(chunk_size), b""):
             md5.update(chunk)
         if md5.hexdigest() != dataset.md5sum:
+            print(
+                f"MD5 checksum mismatch for {filename}: got {md5.hexdigest()} but expected {dataset.md5sum}"
+            )
             raise RuntimeError(f"MD5 checksum mismatch for {filename}")
 
         f.seek(0)
@@ -162,23 +167,50 @@ def _download_extract_dataset(dataset: Dataset, pbar: tqdm):
             tar.extractall(data_directory())
 
 
-def install_datasets(force: bool = False, show_progress: bool = True):
-    os.environ["GEANT4_DATA_DIR"] = data_directory()
+def missing_datasets(directory: str | None = None) -> list[Dataset]:
+    if directory is None:
+        directory = data_directory()
     datasets_to_download = []
+
     for dataset in datasets:
-        path = os.path.join(data_directory(), dataset.name + dataset.version)
-        os.environ[dataset.env] = path
-        if not os.path.exists(path) or force:
+        path = os.path.join(directory, dataset.name + dataset.version)
+        if not os.path.exists(path):
             datasets_to_download.append(dataset)
-
-    if len(datasets_to_download) == 0:
+    return datasets_to_download
+
+
+def check_datasets(throw: bool = False) -> bool:
+    datasets_to_download = missing_datasets()
+    if datasets_to_download:
+        if throw:
+            raise RuntimeError(
+                f"Missing Geant4 datasets: {', '.join([f'{dataset.name}@v{dataset.version}' for dataset in datasets_to_download])}"
+            )
+        return False
+    return True
+
+
+def install_datasets(show_progress: bool = True):
+    # first try to see if the datasets are installed in the application directory
+    datasets_to_download = missing_datasets()
+    if not datasets_to_download:
+        # datasets are installed in application directory
+        os.environ["GEANT4_DATA_DIR"] = data_directory()
         return
 
+    # check if the datasets are present in the corresponding Geant4 directory
+    if not bool(missing_datasets(os.environ["GEANT4_DATA_DIR"])):
+        # return
+        ...
+
+    # download and extract the datasets
+    os.environ["GEANT4_DATA_DIR"] = data_directory()
+
     os.makedirs(data_directory(), exist_ok=True)
     if show_progress:
         print(
             f"""
-Geant4 datasets (<2GB) will be installed to "{data_directory()}".
+Geant4 datasets will be installed to "{data_directory()}".
 This may take a while but only needs to be done once.
 You can override the default location by calling `application_directory(path)` or `application_directory(temp=True)` to use a temporary directory.
 The following Geant4 datasets will be installed: {", ".join([f"{dataset.name}@v{dataset.version}" for dataset in datasets_to_download])}"""
@@ -196,7 +228,7 @@ def install_datasets(force: bool = False, show_progress: bool = True):
         ) as executor:
             futures = [
                 executor.submit(_download_extract_dataset, dataset, pbar)
-                for i, dataset in enumerate(datasets_to_download)
+                for dataset in datasets_to_download
             ]
             concurrent.futures.wait(futures)
 
@@ -212,7 +244,6 @@ def uninstall_datasets():
     package_dir = os.path.dirname(__file__)
 
     if not os.path.relpath(package_dir, dir_to_remove).startswith(".."):
-        # make sure we don't accidentally delete something important
         raise RuntimeError(
             f"Refusing to remove {dir_to_remove} because it is not a subdirectory of {package_dir}"
         )
@@ -221,4 +252,4 @@ def uninstall_datasets():
 
 def reinstall_datasets():
     uninstall_datasets()
-    install_datasets(force=True)
+    install_datasets()