diff --git a/docs/release-notes/0.10.12.md b/docs/release-notes/0.10.12.md
new file mode 100644
index 00000000..40737bed
--- /dev/null
+++ b/docs/release-notes/0.10.12.md
@@ -0,0 +1,11 @@
+### 0.10.12 {small}`the-future`
+
+```{rubric} Features
+```
+```{rubric} Performance
+```
+```{rubric} Bug fixes
+```
+```{rubric} Misc
+```
+* Update `get_random_state` for `scrublet `{pr}`301` {smaller}`S Dicks`
diff --git a/docs/release-notes/index.md b/docs/release-notes/index.md
index faa775ff..1faf4152 100644
--- a/docs/release-notes/index.md
+++ b/docs/release-notes/index.md
@@ -3,6 +3,8 @@
 # Release notes
 
 ## Version 0.10.0
+```{include} /release-notes/0.10.12.md
+```
 ```{include} /release-notes/0.10.11.md
 ```
 ```{include} /release-notes/0.10.10.md
diff --git a/src/rapids_singlecell/preprocessing/_scrublet/core.py b/src/rapids_singlecell/preprocessing/_scrublet/core.py
index b8bb2f0d..372623f9 100644
--- a/src/rapids_singlecell/preprocessing/_scrublet/core.py
+++ b/src/rapids_singlecell/preprocessing/_scrublet/core.py
@@ -9,9 +9,10 @@
 from anndata import AnnData, concat
 from cuml.neighbors import NearestNeighbors
 from cupyx.scipy import sparse
-from scanpy._utils import get_random_state
 from scanpy.preprocessing._utils import sample_comb
 
+from rapids_singlecell.preprocessing._utils import get_random_state
+
 from .sparse_utils import subsample_counts
 
 if TYPE_CHECKING:
diff --git a/src/rapids_singlecell/preprocessing/_scrublet/pipeline.py b/src/rapids_singlecell/preprocessing/_scrublet/pipeline.py
index 9b3779b6..7f52ece0 100644
--- a/src/rapids_singlecell/preprocessing/_scrublet/pipeline.py
+++ b/src/rapids_singlecell/preprocessing/_scrublet/pipeline.py
@@ -3,10 +3,9 @@
 from typing import TYPE_CHECKING, Literal
 
 import cupy as cp
-from cupyx import cusparse
 from cupyx.scipy import sparse
 
-from rapids_singlecell.preprocessing._utils import _get_mean_var
+from rapids_singlecell.preprocessing._utils import _get_mean_var, _sparse_to_dense
 
 from .sparse_utils import sparse_multiply, sparse_zscore
 
@@ -59,10 +58,10 @@ def truncated_svd(
 
     self._counts_obs_norm = self._counts_obs_norm.astype(cp.float32)
     self._counts_sim_norm = self._counts_sim_norm.astype(cp.float32)
-    X_obs = cusparse.sparseToDense(self._counts_obs_norm)
+    X_obs = _sparse_to_dense(self._counts_obs_norm)
     svd = TruncatedSVD(n_components=n_prin_comps, random_state=random_state).fit(X_obs)
     X_obs = svd.transform(X_obs)
-    X_sim = svd.transform(cusparse.sparseToDense(self._counts_sim_norm))
+    X_sim = svd.transform(_sparse_to_dense(self._counts_sim_norm))
     self.set_manifold(X_obs, X_sim)
 
 
@@ -79,8 +78,9 @@ def pca(
 
     self._counts_obs_norm = self._counts_obs_norm.astype(cp.float32)
     self._counts_sim_norm = self._counts_sim_norm.astype(cp.float32)
-    X_obs = cusparse.sparseToDense(self._counts_obs_norm)
+    X_obs = _sparse_to_dense(self._counts_obs_norm)
+
     pca = PCA(n_components=n_prin_comps, random_state=random_state).fit(X_obs)
     X_obs = pca.transform(X_obs)
-    X_sim = pca.transform(cusparse.sparseToDense(self._counts_sim_norm))
+    X_sim = pca.transform(_sparse_to_dense(self._counts_sim_norm))
     self.set_manifold(X_obs, X_sim)
diff --git a/src/rapids_singlecell/preprocessing/_scrublet/sparse_utils.py b/src/rapids_singlecell/preprocessing/_scrublet/sparse_utils.py
index dcbc02c0..eec45a15 100644
--- a/src/rapids_singlecell/preprocessing/_scrublet/sparse_utils.py
+++ b/src/rapids_singlecell/preprocessing/_scrublet/sparse_utils.py
@@ -5,9 +5,8 @@
 import cupy as cp
 import numpy as np
 from cupyx.scipy import sparse
-from scanpy._utils import get_random_state
 
-from rapids_singlecell.preprocessing._utils import _get_mean_var
+from rapids_singlecell.preprocessing._utils import _get_mean_var, get_random_state
 
 if TYPE_CHECKING:
     from numpy.typing import NDArray
diff --git a/src/rapids_singlecell/preprocessing/_utils.py b/src/rapids_singlecell/preprocessing/_utils.py
index 754270fa..1bb533f2 100644
--- a/src/rapids_singlecell/preprocessing/_utils.py
+++ b/src/rapids_singlecell/preprocessing/_utils.py
@@ -4,11 +4,14 @@
 from typing import TYPE_CHECKING, Literal
 
 import cupy as cp
+import numpy as np
 from cupyx.scipy.sparse import issparse, isspmatrix_csc, isspmatrix_csr, spmatrix
 
 if TYPE_CHECKING:
     from anndata import AnnData
 
+    from rapids_singlecell._utils import AnyRandom
+
 
 def _sparse_to_dense(X: spmatrix, order: Literal["C", "F"] | None = None) -> cp.ndarray:
     if order is None:
@@ -154,3 +157,9 @@ def _check_use_raw(adata: AnnData, use_raw: None | bool, layer: str | None) -> b
     if layer is not None:
         return False
     return adata.raw is not None
+
+
+def get_random_state(seed: AnyRandom) -> np.random.RandomState:
+    if isinstance(seed, np.random.RandomState):
+        return seed
+    return np.random.RandomState(seed)