PyVisualFields 2.0.6__tar.gz → 2.0.8__tar.gz

{pyvisualfields-2.0.6/src/PyVisualFields.egg-info → pyvisualfields-2.0.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyVisualFields
-Version: 2.0.6
+Version: 2.0.8
 Summary: A python toolkit for visual field analysis
 Home-page: https://github.com/mohaEs/PyVisualField
 Author: Mohammad Eslami, Bharath Erusalagandi, Mousa Moradi
@@ -77,7 +77,7 @@ The list and description of all functions are available at [All_Functions](#list
 - Data [demo_1_Data.ipynb](demo_1_Data.ipynb)
 - Normalization and deviation analysis [demo_2_Deviation_Analysis.ipynb](demo_2_Deviation_Analysis.ipynb)
 - Plotting [demo_3_Plotting.ipynb](demo_3_Plotting.ipynb)
-- Progression Analysis [demo_4_Analysis.ipynb](demo_4_Analysis.ipynb)
+- Progression Analysis [demo_4_ProgressionAnalysis.ipynb](demo_4_ProgressionAnalysis.ipynb)
 - Glaucoma Detection [demo5_PyGlaucoMetrics.ipynb](demo5_PyGlaucoMetrics.ipynb) </br>
 
 __Notice:__ PyGlaucoMetric is also available as a seperatre PyPI package and GitHub repository (built upon PyVisualFields), which includes a graphical user interface (GUI) for progression analysis and glaucoma detection. Indeed PyVisualFields is designed as a developer-facing package library, while pyGlaucoMetric serves as an accessible GUI application implementing selected visual field analysis components.

{pyvisualfields-2.0.6 → pyvisualfields-2.0.8}/README.md

@@ -38,7 +38,7 @@ The list and description of all functions are available at [All_Functions](#list
 - Data [demo_1_Data.ipynb](demo_1_Data.ipynb)
 - Normalization and deviation analysis [demo_2_Deviation_Analysis.ipynb](demo_2_Deviation_Analysis.ipynb)
 - Plotting [demo_3_Plotting.ipynb](demo_3_Plotting.ipynb)
-- Progression Analysis [demo_4_Analysis.ipynb](demo_4_Analysis.ipynb)
+- Progression Analysis [demo_4_ProgressionAnalysis.ipynb](demo_4_ProgressionAnalysis.ipynb)
 - Glaucoma Detection [demo5_PyGlaucoMetrics.ipynb](demo5_PyGlaucoMetrics.ipynb) </br>
 
 __Notice:__ PyGlaucoMetric is also available as a seperatre PyPI package and GitHub repository (built upon PyVisualFields), which includes a graphical user interface (GUI) for progression analysis and glaucoma detection. Indeed PyVisualFields is designed as a developer-facing package library, while pyGlaucoMetric serves as an accessible GUI application implementing selected visual field analysis components.

{pyvisualfields-2.0.6 → pyvisualfields-2.0.8}/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r", encoding="utf-8") as fh:
 
 setuptools.setup(
     name="PyVisualFields",
-    version="2.0.6",
+    version="2.0.8",
     author="Mohammad Eslami, Bharath Erusalagandi, Mousa Moradi",
     author_email="Mohammad_eslami@meei.harvard.edu",
     description="A python toolkit for visual field analysis",

{pyvisualfields-2.0.6 → pyvisualfields-2.0.8}/src/PyVisualFields/Deviation_Analysis.py

@@ -803,19 +803,111 @@ def py_getgl(df_vf: pd.DataFrame) -> pd.DataFrame:
 
 #     return df_gp
 
+# def py_getglp(df_gi: pd.DataFrame) -> pd.DataFrame:
+#     """Global indices probability values."""
+
+#     nv = _nv()
+#     lut_dict = nv["glp_lut"]
+#     probs = nv["glp_probs"]
+#     idxm_0 = [i - 1 for i in (nv["glp_idxm"] or [])]
+#     idxs_0 = [i - 1 for i in (nv["glp_idxs"] or [])]
+
+#     gl_cols = ["msens", "ssens", "tmd", "tsd", "pmd", "psd", "gh", "vfi"]
+
+#     mean_cols = [gl_cols[i] for i in idxm_0 if i < len(gl_cols)]
+#     std_cols = [gl_cols[i] for i in idxs_0 if i < len(gl_cols)]
+
+#     n_probs = len(probs)
+
+#     prob_vals = np.array([
+#         _parse_lut_value(p) if isinstance(p, str) else float(p)
+#         for p in probs
+#     ])
+
+#     df_gp = df_gi.copy()
+
+#     def _prob_col_name(col):
+#         return f"{col}prob"
+
+#     def _build_col_lut(col):
+#         raw = lut_dict.get(col, [np.nan] * n_probs)
+#         return np.array([
+#             _parse_lut_value(v) if isinstance(v, str) else float(v)
+#             for v in raw
+#         ])
+
+#     # Left-tailed: lower values are abnormal
+#     for col in mean_cols:
+#         if col not in df_gi.columns:
+#             continue
+
+#         cutoffs = _build_col_lut(col)
+#         vals = df_gi[col].values.astype(float)
+#         vp = np.full(len(vals), np.nan)
+
+#         for i in range(n_probs - 1, 0, -1):
+#             vp[vals < cutoffs[i]] = prob_vals[i]
+
+#         df_gp[_prob_col_name(col)] = vp
+
+#     # Right-tailed: higher values are abnormal
+#     for col in std_cols:
+#         if col not in df_gi.columns:
+#             continue
+
+#         cutoffs = _build_col_lut(col)
+#         vals = df_gi[col].values.astype(float)
+#         vp = np.full(len(vals), np.nan)
+
+#         for i in range(n_probs - 1, 0, -1):
+#             vp[vals > cutoffs[i]] = prob_vals[i]
+
+#         df_gp[_prob_col_name(col)] = vp
+
+#     return df_gp
+
+
+
 def py_getglp(df_gi: pd.DataFrame) -> pd.DataFrame:
-    """Global indices probability values."""
+    """Compute missing global-index probability values only."""
 
     nv = _nv()
     lut_dict = nv["glp_lut"]
     probs = nv["glp_probs"]
+
     idxm_0 = [i - 1 for i in (nv["glp_idxm"] or [])]
     idxs_0 = [i - 1 for i in (nv["glp_idxs"] or [])]
 
     gl_cols = ["msens", "ssens", "tmd", "tsd", "pmd", "psd", "gh", "vfi"]
 
     mean_cols = [gl_cols[i] for i in idxm_0 if i < len(gl_cols)]
-    std_cols = [gl_cols[i] for i in idxs_0 if i < len(gl_cols)]
+    std_cols  = [gl_cols[i] for i in idxs_0 if i < len(gl_cols)]
+
+    def _prob_col_name(col):
+        return f"{col}prob"
+
+    # candidates that can be computed
+    computable_cols = [
+        c for c in (mean_cols + std_cols)
+        if c in df_gi.columns
+    ]
+
+    available_gl = [c for c in gl_cols if c in df_gi.columns]
+    missing_gl = [c for c in gl_cols if c not in df_gi.columns]
+
+    print(f"==> py_getglp: available globals: {available_gl}")
+    print(f"==> py_getglp: missing globals: {missing_gl}")
+
+    missing_prob_cols = [
+        c for c in computable_cols
+        if _prob_col_name(c) not in df_gi.columns
+    ]
+
+    print(f"==> py_getglp: missing global probability indices: "
+          f"{[_prob_col_name(c) for c in missing_prob_cols]}")
+
+    if not missing_prob_cols:
+        return df_gi.copy()
 
     n_probs = len(probs)
 
@@ -824,11 +916,6 @@ def py_getglp(df_gi: pd.DataFrame) -> pd.DataFrame:
         for p in probs
     ])
 
-    df_gp = df_gi.copy()
-
-    def _prob_col_name(col):
-        return f"{col}prob"
-
     def _build_col_lut(col):
         raw = lut_dict.get(col, [np.nan] * n_probs)
         return np.array([
@@ -836,9 +923,11 @@ def py_getglp(df_gi: pd.DataFrame) -> pd.DataFrame:
             for v in raw
         ])
 
+    computed = {}
+
     # Left-tailed: lower values are abnormal
     for col in mean_cols:
-        if col not in df_gi.columns:
+        if col not in missing_prob_cols:
             continue
 
         cutoffs = _build_col_lut(col)
@@ -848,11 +937,11 @@ def py_getglp(df_gi: pd.DataFrame) -> pd.DataFrame:
         for i in range(n_probs - 1, 0, -1):
             vp[vals < cutoffs[i]] = prob_vals[i]
 
-        df_gp[_prob_col_name(col)] = vp
+        computed[_prob_col_name(col)] = vp
 
     # Right-tailed: higher values are abnormal
     for col in std_cols:
-        if col not in df_gi.columns:
+        if col not in missing_prob_cols:
             continue
 
         cutoffs = _build_col_lut(col)
@@ -862,10 +951,27 @@ def py_getglp(df_gi: pd.DataFrame) -> pd.DataFrame:
         for i in range(n_probs - 1, 0, -1):
             vp[vals > cutoffs[i]] = prob_vals[i]
 
-        df_gp[_prob_col_name(col)] = vp
+        computed[_prob_col_name(col)] = vp
 
-    return df_gp
+    if "md" in df_gi.columns and "mdprob" not in df_gi.columns:
+        cutoffs = _build_col_lut("tmd")
+        vals = df_gi["md"].values.astype(float)
+        vp = np.full(len(vals), np.nan)
 
+        for i in range(n_probs - 1, 0, -1):
+            vp[vals < cutoffs[i]] = prob_vals[i]
+
+        computed["mdprob"] = vp
+
+    if computed:
+        out = pd.concat(
+            [df_gi.copy(), pd.DataFrame(computed, index=df_gi.index)],
+            axis=1,
+        )
+    else:
+        out = df_gi.copy()
+
+    return out
 
 def py_getallvalues(df_vf: pd.DataFrame):
     """Compute all deviations and global indices in one call.

pyvisualfields-2.0.8/src/PyVisualFields/__init__.py

@@ -0,0 +1 @@
+__version__ = "2.0.8"

{pyvisualfields-2.0.6 → pyvisualfields-2.0.8}/src/PyVisualFields/utils.py

@@ -119,34 +119,62 @@ META_ALIASES = {
     "vfi": [
         "vfi"
     ],
+    #probability
+    "vfiprob": [
+        "vfi_p", "vfiprob", "vfip"
+    ],
 
     "msens": [
         "msens", "ms", "meansensitivity"
     ],
+    #probability
+    "msensprob": [
+        "msens_p", "msensprob", "msenp", "ms_p", "msprob", "msp"
+    ],
 
     "ssens": [
         "ssens", "ss", "sdsensitivity"
     ],
+    #probability
+    "ssensprob": [
+        "ssens_p", "ssensprob", "lsenp", "ss_p", "ssprob", "ssp"
+    ],
 
     "tmd": [
         "tmd"
     ],
+    #probability
+    "tmdprob": [
+        "tmd_p", "tmdprob", "tmdp"
+    ],
 
     "tsd": [
         "tsd"
     ],
+    #probability
+    "tsdprob": [
+        "tsd_p", "tsdprob", "tsdp"
+    ],
 
     "pmd": [
         "pmd"
     ],
+    #probability
+    "pmdprob": [
+        "pmd_p", "pmdprob", "pmdp"
+    ],
 
     "gh": [
         "gh", "generalheight"
     ],
+    #probability
+    "ghprob": [
+        "gh_p", "ghprob", "ghp"
+    ],
 
     # Reliability
     "fpr": [
-        "fpr"
+        "fpr" 
     ],
 
     "fnr": [

{pyvisualfields-2.0.6 → pyvisualfields-2.0.8/src/PyVisualFields.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyVisualFields
-Version: 2.0.6
+Version: 2.0.8
 Summary: A python toolkit for visual field analysis
 Home-page: https://github.com/mohaEs/PyVisualField
 Author: Mohammad Eslami, Bharath Erusalagandi, Mousa Moradi
@@ -77,7 +77,7 @@ The list and description of all functions are available at [All_Functions](#list
 - Data [demo_1_Data.ipynb](demo_1_Data.ipynb)
 - Normalization and deviation analysis [demo_2_Deviation_Analysis.ipynb](demo_2_Deviation_Analysis.ipynb)
 - Plotting [demo_3_Plotting.ipynb](demo_3_Plotting.ipynb)
-- Progression Analysis [demo_4_Analysis.ipynb](demo_4_Analysis.ipynb)
+- Progression Analysis [demo_4_ProgressionAnalysis.ipynb](demo_4_ProgressionAnalysis.ipynb)
 - Glaucoma Detection [demo5_PyGlaucoMetrics.ipynb](demo5_PyGlaucoMetrics.ipynb) </br>
 
 __Notice:__ PyGlaucoMetric is also available as a seperatre PyPI package and GitHub repository (built upon PyVisualFields), which includes a graphical user interface (GUI) for progression analysis and glaucoma detection. Indeed PyVisualFields is designed as a developer-facing package library, while pyGlaucoMetric serves as an accessible GUI application implementing selected visual field analysis components.

pyvisualfields-2.0.6/src/PyVisualFields/__init__.py

@@ -1 +0,0 @@
-__version__ = "2.0.6"