Breast　cancer　is　the　2nd　leading　cancer　of　death　among　women　around　the　world.　In　Asia　and　Africa　due　to　low　income,　the　mortality　rates　are　very　high　as　compared　to　Europe　and　America.　Initially,　image　interpretation　is　manually　conducted　by　the　radiologist　and　physicians　that　requires　expertise;　thus,　the　computer-aided　diagnostic　is　necessary　to　enhance　the　accuracy　of　cancer　diagnostics　in　mammograms　at　early　stages.　To　overcome　human　error　computer-aided　system　was　developed　based　on　machine　learning　and　deep　learning　algorithm　to　process　medical　images　with　efficient　accuracy　for　the　diagnosis　of　cancer　and　assist　the　physician　for　better　decisions　making.　This　research　aims　to　present　the　state-of-the-art　machine　learning　techniques　for　the　detection　of　breast　cancer,　and　critically　analysis　of　the　current　literature　in　this　area　to　identify　the　research　gap.　There　are　many　studies　presented　in　the　literature　to　achieve　similar　goals.　The　main　difference　between　these　studies　and　this　review　is　that　this　paper　is　more　focused　on　those　modalities　that　can　figure　out　breast　composition,　mass,　density,　calcification,　and　architectural　distortion.　This　study　includes　a　summary　of　110　papers,　pointing　out　which　techniques　are　applied　for　image　preprocessing　and　classification,　which　method　is　implemented　for　the　detection　of　breast　density,　mass,　and　calcification　from　mammogram　images.　Furthermore,　we　critically　analyzed　the　performance　measuring　parameters　for　the　evaluation　of　results　and　the　datasets　that　have　been　used　for　experiments.　Another　focus　in　this　review　is　to　assess　the　modalities　and　features　that　can　be　helpful　for　the　assessment　of　grading　in　mammogram　images.