. 2018 Aug 15;154(10):1159–1166. doi: 10.1001/jamadermatol.2018.2388

Table 2. Agreement in Skin Pathology Interpretation and Assessment for Clinical Discrepancy Between TM and Digital WSI in 5 Sets of Melanocytic Lesion Cases.

Reference Diagnosis^a	I	II	III	IV	V	Total No. of Interpretations	Agreement of Pathologists With Consensus Reference, % (95% CI)
Participating pathologists' interpretation with TM
I	125^b	7	5	0	1	138	91 (84-95)
II	170	64^b	32	6	4	276	23 (18-30)
III	109	80	201^b	21^c	3^c	414	49 (43-54)
IV	18^d	37^d	119^d	198^b	42	414	48 (41-55)
V	9^e	6^e	41^e	61^e	297^b	414	72 (67-76)
Participating pathologists' interpretation with WSI
I	103^b	11	7	2	0	123	84 (75-90)
II	137	60^b	32	11	6	246	24 (19-31)
III	110	77	143^b	38^c	1^c	369	39 (31-47)
IV	34^d	30^d	98^d	176^b	31	369	48 (39-56)
V	11^e	8^e	32^e	53^e	265^b	369	72 (66-77)

Abbreviations: TM, traditional microscopy; WSI, whole-slide imaging.

^{^a}

Consensus Melanocytic Pathology Assessment Tool and Hierarchy for Diagnosis (MPATH-Dx) reference diagnosis was obtained using the TM format; roman numerals indicate MPATH-Dx classes, with I indicating the least severe and V the most severe.

^{^b}

Reference diagnosis is concordant with participating pathologists’ interpretation.

^{^c}

Reference diagnosis is class III; participating pathologists' interpretation as class IV or V: traditional 6%, digital 11%.

^{^d}

Reference diagnosis is class IV; participating pathologists' interpretation as class I, II, or III: traditional 42%, digital 44%.

^{^e}

Reference diagnosis is class V; participating pathologists' interpretation as class I, II, III, or IV: traditional 28%, digital 28%.